1PCS(8) System Administration Utilities PCS(8)
2
6 pcs - pacemaker/corosync configuration system
7
9 pcs [-f file] [-h] [commands]...
10
12 Control and configure pacemaker and corosync.
13
15 -h, --help
16 Display usage and exit.
17 -f file
19 Perform actions on file instead of active CIB.
20 Commands supporting the option use the initial state of the
21 specified file as their input and then overwrite the file with
22 the state reflecting the requested operation(s).
23 A few commands only use the specified file in read-only mode
24 since their effect is not a CIB modification.
25 --debug
27 Print all network traffic and external commands run.
28 --version
30 Print pcs version information. List pcs capabilities if --full
31 is specified.
32 --request-timeout=<timeout>
34 Timeout for each outgoing request to another node in seconds.
35 Default is 60s.
36 Commands:
38 cluster
39 Configure cluster options and nodes.
40 resource
42 Manage cluster resources.
43 stonith
45 Manage fence devices.
46 constraint
48 Manage resource constraints.
49 property
51 Manage pacemaker properties.
52 acl
54 Manage pacemaker access control lists.
55 qdevice
57 Manage quorum device provider on the local host.
58 quorum
60 Manage cluster quorum settings.
61 booth
63 Manage booth (cluster ticket manager).
64 status
66 View cluster status.
67 config
69 View and manage cluster configuration.
70 pcsd
72 Manage pcs daemon.
73 host
75 Manage hosts known to pcs/pcsd.
76 node
78 Manage cluster nodes.
79 alert
81 Manage pacemaker alerts.
82 client
84 Manage pcsd client configuration.
85 dr
87 Manage disaster recovery configuration.
88 tag
90 Manage pacemaker tags.
91 resource
93 [status [<resource id | tag id>] [node=<node>] [--hide-inactive]]
94 Show status of all currently configured resources. If --hide-in‐
95 active is specified, only show active resources. If a resource
96 or tag id is specified, only show status of the specified re‐
97 source or resources in the specified tag. If node is specified,
98 only show status of resources configured for the specified node.
99 config [--output-format text|cmd|json] [<resource id>]...
101 Show options of all currently configured resources or if re‐
102 source ids are specified show the options for the specified re‐
103 source ids. There are 3 formats of output available: 'cmd',
104 'json' and 'text', default is 'text'. Format 'text' is a human
105 friendly output. Format 'cmd' prints pcs commands which can be
106 used to recreate the same configuration. Format 'json' is a ma‐
107 chine oriented output of the configuration.
108 list [filter] [--nodesc]
110 Show list of all available resource agents (if filter is pro‐
111 vided then only resource agents matching the filter will be
112 shown). If --nodesc is used then descriptions of resource agents
113 are not printed.
114 describe [<standard>:[<provider>:]]<type> [--full]
116 Show options for the specified resource. If --full is specified,
117 all options including advanced and deprecated ones are shown.
118 create <resource id> [<standard>:[<provider>:]]<type> [<resource op‐
120 tions>] [op <operation action> <operation options> [<operation action>
121 <operation options>]...] [meta <meta options>] [clone [<clone id>]
122 [<clone meta options>] | promotable [<clone id>] [<promotable meta op‐
123 tions>] | --group <group id> [--before <resource id> | --after <re‐
124 source id>] | bundle <bundle id>] [--disabled] [--agent-validation]
125 [--no-default-ops] [--wait[=n]]
126 create --future <resource id> [<standard>:[<provider>:]]<type> [<re‐
127 source options>] [op <operation action> <operation options> [<operation
128 action> <operation options>]...] [meta <meta options>] [clone [<clone
129 id>] [meta <clone meta options>] | promotable [<clone id>] [meta <pro‐
130 motable meta options>] | --group <group id> [--before <resource id> |
131 --after <resource id>] | bundle <bundle id>] [--disabled] [--agent-val‐
132 idation] [--no-default-ops] [--wait[=n]]
133 Create specified resource.
134 If clone is used, a clone resource is created. If promotable is
135 used, a promotable clone resource is created.
136 If --group is specified, the resource is added to the group
137 named. You can use --before or --after to specify the position
138 of the added resource relatively to some resource already exist‐
139 ing in the group.
140 If bundle is specified, resource will be created inside of the
141 specified bundle.
142 If --disabled is specified, the resource is not started automat‐
143 ically.
144 If --agent-validation is specified, resource agent validate-all
145 action will be used to validate resource options.
146 If --no-default-ops is specified, only monitor operations are
147 created for the resource and all other operations use default
148 settings.
149 If --wait is specified, pcs will wait up to 'n' seconds for the
150 resource to start and then return 0 if the resource is started,
151 or 1 if the resource has not yet started. If 'n' is not speci‐
152 fied it defaults to 60 minutes.
153 Specifying --future switches to the new command format that
154 changes the way clone and promotable meta options are expected
155 to be specified.
156 Example: Create a new resource called 'VirtualIP' with IP ad‐
158 dress 192.168.0.99, netmask of 32, monitored everything 30 sec‐
159 onds, on eth2:
160 pcs resource create VirtualIP ocf:heartbeat:IPaddr2
161 ip=192.168.0.99 cidr_netmask=32 nic=eth2 op monitor interval=30s
162 delete <resource id|group id|bundle id|clone id>
164 Deletes the resource, group, bundle or clone (and all resources
165 within the group/bundle/clone).
166 remove <resource id|group id|bundle id|clone id>
168 Deletes the resource, group, bundle or clone (and all resources
169 within the group/bundle/clone).
170 enable <resource id | tag id>... [--wait[=n]]
172 Allow the cluster to start the resources. Depending on the rest
173 of the configuration (constraints, options, failures, etc), the
174 resources may remain stopped. If --wait is specified, pcs will
175 wait up to 'n' seconds for the resources to start and then re‐
176 turn 0 if the resources are started, or 1 if the resources have
177 not yet started. If 'n' is not specified it defaults to 60 min‐
178 utes.
179 disable <resource id | tag id>... [--safe [--brief] [--no-strict]]
181 [--simulate [--brief]] [--wait[=n]]
182 Attempt to stop the resources if they are running and forbid the
183 cluster from starting them again. Depending on the rest of the
184 configuration (constraints, options, failures, etc), the re‐
185 sources may remain started.
186 If --safe is specified, no changes to the cluster configuration
187 will be made if other than specified resources would be affected
188 in any way. If --brief is also specified, only errors are
189 printed.
190 If --no-strict is specified, no changes to the cluster configu‐
191 ration will be made if other than specified resources would get
192 stopped or demoted. Moving resources between nodes is allowed.
193 If --simulate is specified, no changes to the cluster configura‐
194 tion will be made and the effect of the changes will be printed
195 instead. If --brief is also specified, only a list of affected
196 resources not specified in the command will be printed.
197 If --wait is specified, pcs will wait up to 'n' seconds for the
198 resources to stop and then return 0 if the resources are stopped
199 or 1 if the resources have not stopped. If 'n' is not specified
200 it defaults to 60 minutes.
201 safe-disable <resource id | tag id>... [--brief] [--no-strict] [--simu‐
203 late [--brief]] [--wait[=n]] [--force]
204 Attempt to stop the resources if they are running and forbid the
205 cluster from starting them again. Depending on the rest of the
206 configuration (constraints, options, failures, etc), the re‐
207 sources may remain started. No changes to the cluster configura‐
208 tion will be made if other than specified resources would be af‐
209 fected in any way.
210 If --brief is specified, only errors are printed.
211 If --no-strict is specified, no changes to the cluster configu‐
212 ration will be made if other than specified resources would get
213 stopped or demoted. Moving resources between nodes is allowed.
214 If --simulate is specified, no changes to the cluster configura‐
215 tion will be made and the effect of the changes will be printed
216 instead. If --brief is also specified, only a list of affected
217 resources not specified in the command will be printed.
218 If --wait is specified, pcs will wait up to 'n' seconds for the
219 resources to stop and then return 0 if the resources are stopped
220 or 1 if the resources have not stopped. If 'n' is not specified
221 it defaults to 60 minutes.
222 If --force is specified, checks for safe disable will be
223 skipped.
224 restart <resource id> [node] [--wait=n]
226 Restart the resource specified. If a node is specified and if
227 the resource is a clone or bundle it will be restarted only on
228 the node specified. If --wait is specified, then we will wait up
229 to 'n' seconds for the resource to be restarted and return 0 if
230 the restart was successful or 1 if it was not.
231 debug-start <resource id> [--full]
233 This command will force the specified resource to start on this
234 node ignoring the cluster recommendations and print the output
235 from starting the resource. Using --full will give more de‐
236 tailed output. This is mainly used for debugging resources that
237 fail to start.
238 debug-stop <resource id> [--full]
240 This command will force the specified resource to stop on this
241 node ignoring the cluster recommendations and print the output
242 from stopping the resource. Using --full will give more de‐
243 tailed output. This is mainly used for debugging resources that
244 fail to stop.
245 debug-promote <resource id> [--full]
247 This command will force the specified resource to be promoted on
248 this node ignoring the cluster recommendations and print the
249 output from promoting the resource. Using --full will give more
250 detailed output. This is mainly used for debugging resources
251 that fail to promote.
252 debug-demote <resource id> [--full]
254 This command will force the specified resource to be demoted on
255 this node ignoring the cluster recommendations and print the
256 output from demoting the resource. Using --full will give more
257 detailed output. This is mainly used for debugging resources
258 that fail to demote.
259 debug-monitor <resource id> [--full]
261 This command will force the specified resource to be monitored
262 on this node ignoring the cluster recommendations and print the
263 output from monitoring the resource. Using --full will give
264 more detailed output. This is mainly used for debugging re‐
265 sources that fail to be monitored.
266 move <resource id> [destination node] [--promoted] [--strict]
268 [--wait[=n]]
269 Move the resource off the node it is currently running on. This
270 is achieved by creating a -INFINITY location constraint to ban
271 the node. If destination node is specified the resource will be
272 moved to that node by creating an INFINITY location constraint
273 to prefer the destination node. The constraint needed for moving
274 the resource will be automatically removed once the resource is
275 running on it's new location. The command will fail in case it
276 is not possible to verify that the resource will not be moved
277 back after deleting the constraint.
278 If --strict is specified, the command will also fail if other
280 resources would be affected.
281 If --promoted is used the scope of the command is limited to the
283 Promoted role and promotable clone id must be used (instead of
284 the resource id).
285 If --wait is specified, pcs will wait up to 'n' seconds for the
287 resource to move and then return 0 on success or 1 on error. If
288 'n' is not specified it defaults to 60 minutes.
289 NOTE: This command has been changed in pcs-0.11. It is equiva‐
291 lent to command 'resource move <resource id> --autodelete' from
292 pcs-0.10.9. Legacy functionality of the 'resource move' command
293 is still available as 'resource move-with-constraint <resource
294 id>'.
295 If you want the resource to preferably avoid running on some
297 nodes but be able to failover to them use 'pcs constraint loca‐
298 tion avoids'.
299 move-with-constraint <resource id> [destination node] [lifetime=<life‐
301 time>] [--promoted] [--wait[=n]]
302 Move the resource off the node it is currently running on by
303 creating a -INFINITY location constraint to ban the node. If
304 destination node is specified the resource will be moved to that
305 node by creating an INFINITY location constraint to prefer the
306 destination node.
307 If lifetime is specified then the constraint will expire after
309 that time, otherwise it defaults to infinity and the constraint
310 can be cleared manually with 'pcs resource clear' or 'pcs con‐
311 straint delete'. Lifetime is expected to be specified as ISO
312 8601 duration (see https://en.wikipedia.org/wiki/ISO_8601#Dura‐
313 tions).
314 If --promoted is used the scope of the command is limited to the
316 Promoted role and promotable clone id must be used (instead of
317 the resource id).
318 If --wait is specified, pcs will wait up to 'n' seconds for the
320 resource to move and then return 0 on success or 1 on error. If
321 'n' is not specified it defaults to 60 minutes.
322 If you want the resource to preferably avoid running on some
324 nodes but be able to failover to them use 'pcs constraint loca‐
325 tion avoids'.
326 ban <resource id> [node] [--promoted] [lifetime=<lifetime>]
328 [--wait[=n]]
329 Prevent the resource id specified from running on the node (or
330 on the current node it is running on if no node is specified) by
331 creating a -INFINITY location constraint.
332 If --promoted is used the scope of the command is limited to the
334 Promoted role and promotable clone id must be used (instead of
335 the resource id).
336 If lifetime is specified then the constraint will expire after
338 that time, otherwise it defaults to infinity and the constraint
339 can be cleared manually with 'pcs resource clear' or 'pcs con‐
340 straint delete'. Lifetime is expected to be specified as ISO
341 8601 duration (see https://en.wikipedia.org/wiki/ISO_8601#Dura‐
342 tions).
343 If --wait is specified, pcs will wait up to 'n' seconds for the
345 resource to move and then return 0 on success or 1 on error. If
346 'n' is not specified it defaults to 60 minutes.
347 If you want the resource to preferably avoid running on some
349 nodes but be able to failover to them use 'pcs constraint loca‐
350 tion avoids'.
351 clear <resource id> [node] [--promoted] [--expired] [--wait[=n]]
353 Remove constraints created by move and/or ban on the specified
354 resource (and node if specified).
355 If --promoted is used the scope of the command is limited to the
357 Promoted role and promotable clone id must be used (instead of
358 the resource id).
359 If --expired is specified, only constraints with expired life‐
361 times will be removed.
362 If --wait is specified, pcs will wait up to 'n' seconds for the
364 operation to finish (including starting and/or moving resources
365 if appropriate) and then return 0 on success or 1 on error. If
366 'n' is not specified it defaults to 60 minutes.
367 standards
369 List available resource agent standards supported by this in‐
370 stallation (OCF, LSB, etc.).
371 providers
373 List available OCF resource agent providers.
374 agents [standard[:provider]]
376 List available agents optionally filtered by standard and
377 provider.
378 update <resource id> [resource options] [op [<operation action> <opera‐
380 tion options>]...] [meta <meta options>] [--agent-validation]
381 [--wait[=n]]
382 Add, remove or change options of specified resource, clone or
383 multi-state resource. Unspecified options will be kept un‐
384 changed. If you wish to remove an option, set it to empty value,
385 i.e. 'option_name='.
386 If an operation (op) is specified it will update the first found
388 operation with the same action on the specified resource. If no
389 operation with that action exists then a new operation will be
390 created. (WARNING: all existing options on the updated operation
391 will be reset if not specified.) If you want to create multiple
392 monitor operations you should use the 'op add' & 'op remove'
393 commands.
394 If --agent-validation is specified, resource agent validate-all
396 action will be used to validate resource options.
397 If --wait is specified, pcs will wait up to 'n' seconds for the
399 changes to take effect and then return 0 if the changes have
400 been processed or 1 otherwise. If 'n' is not specified it de‐
401 faults to 60 minutes.
402 op add <resource id> <operation action> [operation properties]
404 Add operation for specified resource.
405 op delete <resource id> <operation action> [<operation properties>...]
407 Remove specified operation (note: you must specify the exact op‐
408 eration properties to properly remove an existing operation).
409 op delete <operation id>
411 Remove the specified operation id.
412 op remove <resource id> <operation action> [<operation properties>...]
414 Remove specified operation (note: you must specify the exact op‐
415 eration properties to properly remove an existing operation).
416 op remove <operation id>
418 Remove the specified operation id.
419 op defaults [config] [--all] [--full] [--no-check-expired]
421 List currently configured default values for operations. If
422 --all is specified, also list expired sets of values. If --full
423 is specified, also list ids. If --no-expire-check is specified,
424 do not evaluate whether sets of values are expired.
425 op defaults <name>=<value>...
427 Set default values for operations.
428 NOTE: Defaults do not apply to resources / stonith devices which
429 override them with their own defined values.
430 op defaults set create [<set options>] [meta [<name>=<value>]...] [rule
432 [<expression>]]
433 Create a new set of default values for resource / stonith device
434 operations. You may specify a rule describing resources /
435 stonith devices and / or operations to which the set applies.
436 Set options are: id, score
438 Expression looks like one of the following:
440 op <operation name> [interval=<interval>]
441 resource [<standard>]:[<provider>]:[<type>]
442 defined|not_defined <node attribute>
443 <node attribute> lt|gt|lte|gte|eq|ne [string|integer|num‐
444 ber|version] <value>
445 date gt|lt <date>
446 date in_range [<date>] to <date>
447 date in_range <date> to duration <duration options>
448 date-spec <date-spec options>
449 <expression> and|or <expression>
450 (<expression>)
451 You may specify all or any of 'standard', 'provider' and 'type'
453 in a resource expression. For example: 'resource ocf::' matches
454 all resources of 'ocf' standard, while 'resource ::Dummy'
455 matches all resources of 'Dummy' type regardless of their stan‐
456 dard and provider.
457 Dates are expected to conform to ISO 8601 format.
459 Duration options are: hours, monthdays, weekdays, yearsdays,
461 months, weeks, years, weekyears, moon. Value for these options
462 is an integer.
463 Date-spec options are: hours, monthdays, weekdays, yearsdays,
465 months, weeks, years, weekyears, moon. Value for these options
466 is an integer or a range written as integer-integer.
467 NOTE: Defaults do not apply to resources / stonith devices which
469 override them with their own defined values.
470 op defaults set delete [<set id>]...
472 Delete specified options sets.
473 op defaults set remove [<set id>]...
475 Delete specified options sets.
476 op defaults set update <set id> [meta [<name>=<value>]...]
478 Add, remove or change values in specified set of default values
479 for resource / stonith device operations. Unspecified options
480 will be kept unchanged. If you wish to remove an option, set it
481 to empty value, i.e. 'option_name='.
482 NOTE: Defaults do not apply to resources / stonith devices which
484 override them with their own defined values.
485 op defaults update <name>=<value>...
487 Add, remove or change default values for operations. This is a
488 simplified command useful for cases when you only manage one set
489 of default values. Unspecified options will be kept unchanged.
490 If you wish to remove an option, set it to empty value, i.e.
491 'option_name='.
492 NOTE: Defaults do not apply to resources / stonith devices which
494 override them with their own defined values.
495 meta <resource id | group id | clone id> <meta options> [--wait[=n]]
497 Add specified options to the specified resource, group or clone.
498 Meta options should be in the format of name=value, options may
499 be removed by setting an option without a value. If --wait is
500 specified, pcs will wait up to 'n' seconds for the changes to
501 take effect and then return 0 if the changes have been processed
502 or 1 otherwise. If 'n' is not specified it defaults to 60 min‐
503 utes.
504 Example: pcs resource meta TestResource failure-timeout=50 re‐
505 source-stickiness=
506 group list
508 Show all currently configured resource groups and their re‐
509 sources.
510 group add <group id> <resource id> [resource id] ... [resource id]
512 [--before <resource id> | --after <resource id>] [--wait[=n]]
513 Add the specified resource to the group, creating the group if
514 it does not exist. If the resource is present in another group
515 it is moved to the new group. If the group remains empty after
516 move, it is deleted (for cloned groups, the clone is deleted as
517 well). The delete operation may fail in case the group is refer‐
518 enced within the configuration, e.g. by constraints. In that
519 case, use 'pcs resource ungroup' command prior to moving all re‐
520 sources out of the group.
521 You can use --before or --after to specify the position of the
523 added resources relatively to some resource already existing in
524 the group. By adding resources to a group they are already in
525 and specifying --after or --before you can move the resources in
526 the group.
527 If --wait is specified, pcs will wait up to 'n' seconds for the
529 operation to finish (including moving resources if appropriate)
530 and then return 0 on success or 1 on error. If 'n' is not speci‐
531 fied it defaults to 60 minutes.
532 group delete <group id> [resource id]... [--wait[=n]]
534 Remove the group (note: this does not remove any resources from
535 the cluster) or if resources are specified, remove the specified
536 resources from the group. If --wait is specified, pcs will wait
537 up to 'n' seconds for the operation to finish (including moving
538 resources if appropriate) and the return 0 on success or 1 on
539 error. If 'n' is not specified it defaults to 60 minutes.
540 group remove <group id> [resource id]... [--wait[=n]]
542 Remove the group (note: this does not remove any resources from
543 the cluster) or if resources are specified, remove the specified
544 resources from the group. If --wait is specified, pcs will wait
545 up to 'n' seconds for the operation to finish (including moving
546 resources if appropriate) and the return 0 on success or 1 on
547 error. If 'n' is not specified it defaults to 60 minutes.
548 ungroup <group id> [resource id]... [--wait[=n]]
550 Remove the group (note: this does not remove any resources from
551 the cluster) or if resources are specified, remove the specified
552 resources from the group. If --wait is specified, pcs will wait
553 up to 'n' seconds for the operation to finish (including moving
554 resources if appropriate) and the return 0 on success or 1 on
555 error. If 'n' is not specified it defaults to 60 minutes.
556 clone <resource id | group id> [<clone id>] [meta <clone meta options>]
558 [--wait[=n]]
559 Set up the specified resource or group as a clone. If --wait is
560 specified, pcs will wait up to 'n' seconds for the operation to
561 finish (including starting clone instances if appropriate) and
562 then return 0 on success or 1 on error. If 'n' is not specified
563 it defaults to 60 minutes.
564 promotable <resource id | group id> [<clone id>] [meta <clone meta op‐
566 tions>] [--wait[=n]]
567 Set up the specified resource or group as a promotable clone.
568 This is an alias for 'pcs resource clone <resource id> pro‐
569 motable=true'.
570 unclone <clone id | resource id | group id> [--wait[=n]]
572 Remove the specified clone or the clone which contains the spec‐
573 ified group or resource (the resource or group will not be re‐
574 moved). If --wait is specified, pcs will wait up to 'n' seconds
575 for the operation to finish (including stopping clone instances
576 if appropriate) and then return 0 on success or 1 on error. If
577 'n' is not specified it defaults to 60 minutes.
578 bundle create <bundle id> container <container type> [<container op‐
580 tions>] [network <network options>] [port-map <port options>]... [stor‐
581 age-map <storage options>]... [meta <meta options>] [--disabled]
582 [--wait[=n]]
583 Create a new bundle encapsulating no resources. The bundle can
584 be used either as it is or a resource may be put into it at any
585 time. If --disabled is specified, the bundle is not started au‐
586 tomatically. If --wait is specified, pcs will wait up to 'n'
587 seconds for the bundle to start and then return 0 on success or
588 1 on error. If 'n' is not specified it defaults to 60 minutes.
589 bundle reset <bundle id> [container <container options>] [network <net‐
591 work options>] [port-map <port options>]... [storage-map <storage op‐
592 tions>]... [meta <meta options>] [--disabled] [--wait[=n]]
593 Configure specified bundle with given options. Unlike bundle up‐
594 date, this command resets the bundle according given options -
595 no previous options are kept. Resources inside the bundle are
596 kept as they are. If --disabled is specified, the bundle is not
597 started automatically. If --wait is specified, pcs will wait up
598 to 'n' seconds for the bundle to start and then return 0 on suc‐
599 cess or 1 on error. If 'n' is not specified it defaults to 60
600 minutes.
601 bundle update <bundle id> [container <container options>] [network
603 <network options>] [port-map (add <port options>) | (delete | remove
604 <id>...)]... [storage-map (add <storage options>) | (delete | remove
605 <id>...)]... [meta <meta options>] [--wait[=n]]
606 Add, remove or change options of specified bundle. Unspecified
607 options will be kept unchanged. If you wish to remove an option,
608 set it to empty value, i.e. 'option_name='.
609 If you wish to update a resource encapsulated in the bundle, use
611 the 'pcs resource update' command instead and specify the re‐
612 source id.
613 If --wait is specified, pcs will wait up to 'n' seconds for the
615 operation to finish (including moving resources if appropriate)
616 and then return 0 on success or 1 on error. If 'n' is not speci‐
617 fied it defaults to 60 minutes.
618 manage <resource id | tag id>... [--monitor]
620 Set resources listed to managed mode (default). If --monitor is
621 specified, enable all monitor operations of the resources.
622 unmanage <resource id | tag id>... [--monitor]
624 Set resources listed to unmanaged mode. When a resource is in
625 unmanaged mode, the cluster is not allowed to start nor stop the
626 resource. If --monitor is specified, disable all monitor opera‐
627 tions of the resources.
628 defaults [config] [--all] [--full] [--no-check-expired]
630 List currently configured default values for resources / stonith
631 devices. If --all is specified, also list expired sets of val‐
632 ues. If --full is specified, also list ids. If --no-expire-check
633 is specified, do not evaluate whether sets of values are ex‐
634 pired.
635 defaults <name>=<value>...
637 Set default values for resources / stonith devices.
638 NOTE: Defaults do not apply to resources / stonith devices which
639 override them with their own defined values.
640 defaults set create [<set options>] [meta [<name>=<value>]...] [rule
642 [<expression>]]
643 Create a new set of default values for resources / stonith de‐
644 vices. You may specify a rule describing resources / stonith de‐
645 vices to which the set applies.
646 Set options are: id, score
648 Expression looks like one of the following:
650 resource [<standard>]:[<provider>]:[<type>]
651 date gt|lt <date>
652 date in_range [<date>] to <date>
653 date in_range <date> to duration <duration options>
654 date-spec <date-spec options>
655 <expression> and|or <expression>
656 (<expression>)
657 You may specify all or any of 'standard', 'provider' and 'type'
659 in a resource expression. For example: 'resource ocf::' matches
660 all resources of 'ocf' standard, while 'resource ::Dummy'
661 matches all resources of 'Dummy' type regardless of their stan‐
662 dard and provider.
663 Dates are expected to conform to ISO 8601 format.
665 Duration options are: hours, monthdays, weekdays, yearsdays,
667 months, weeks, years, weekyears, moon. Value for these options
668 is an integer.
669 Date-spec options are: hours, monthdays, weekdays, yearsdays,
671 months, weeks, years, weekyears, moon. Value for these options
672 is an integer or a range written as integer-integer.
673 NOTE: Defaults do not apply to resources / stonith devices which
675 override them with their own defined values.
676 defaults set delete [<set id>]...
678 Delete specified options sets.
679 defaults set remove [<set id>]...
681 Delete specified options sets.
682 defaults set update <set id> [meta [<name>=<value>]...]
684 Add, remove or change values in specified set of default values
685 for resources / stonith devices. Unspecified options will be
686 kept unchanged. If you wish to remove an option, set it to empty
687 value, i.e. 'option_name='.
688 NOTE: Defaults do not apply to resources / stonith devices which
690 override them with their own defined values.
691 defaults update <name>=<value>...
693 Add, remove or change default values for resources / stonith de‐
694 vices. This is a simplified command useful for cases when you
695 only manage one set of default values. Unspecified options will
696 be kept unchanged. If you wish to remove an option, set it to
697 empty value, i.e. 'option_name='.
698 NOTE: Defaults do not apply to resources / stonith devices which
700 override them with their own defined values.
701 cleanup [<resource id | stonith id>] [node=<node>] [operation=<opera‐
703 tion> [interval=<interval>]] [--strict]
704 Make the cluster forget failed operations from history of the
705 resource / stonith device and re-detect its current state. This
706 can be useful to purge knowledge of past failures that have
707 since been resolved.
708 If the named resource is part of a group, or one numbered in‐
710 stance of a clone or bundled resource, the clean-up applies to
711 the whole collective resource unless --strict is given.
712 If a resource id / stonith id is not specified then all re‐
714 sources / stonith devices will be cleaned up.
715 If a node is not specified then resources / stonith devices on
717 all nodes will be cleaned up.
718 refresh [<resource id | stonith id>] [node=<node>] [--strict]
720 Make the cluster forget the complete operation history (includ‐
721 ing failures) of the resource / stonith device and re-detect its
722 current state. If you are interested in forgetting failed opera‐
723 tions only, use the 'pcs resource cleanup' command.
724 If the named resource is part of a group, or one numbered in‐
726 stance of a clone or bundled resource, the refresh applies to
727 the whole collective resource unless --strict is given.
728 If a resource id / stonith id is not specified then all re‐
730 sources / stonith devices will be refreshed.
731 If a node is not specified then resources / stonith devices on
733 all nodes will be refreshed.
734 failcount [show [<resource id | stonith id>] [node=<node>] [opera‐
736 tion=<operation> [interval=<interval>]]] [--full]
737 Show current failcount for resources and stonith devices, op‐
738 tionally filtered by a resource / stonith device, node, opera‐
739 tion and its interval. If --full is specified do not sum fail‐
740 counts per resource / stonith device and node. Use 'pcs resource
741 cleanup' or 'pcs resource refresh' to reset failcounts.
742 relocate dry-run [resource1] [resource2] ...
744 The same as 'relocate run' but has no effect on the cluster.
745 relocate run [resource1] [resource2] ...
747 Relocate specified resources to their preferred nodes. If no
748 resources are specified, relocate all resources. This command
749 calculates the preferred node for each resource while ignoring
750 resource stickiness. Then it creates location constraints which
751 will cause the resources to move to their preferred nodes. Once
752 the resources have been moved the constraints are deleted auto‐
753 matically. Note that the preferred node is calculated based on
754 current cluster status, constraints, location of resources and
755 other settings and thus it might change over time.
756 relocate show
758 Display current status of resources and their optimal node ig‐
759 noring resource stickiness.
760 relocate clear
762 Remove all constraints created by the 'relocate run' command.
763 utilization [<resource id> [<name>=<value> ...]]
765 Add specified utilization options to specified resource. If re‐
766 source is not specified, shows utilization of all resources. If
767 utilization options are not specified, shows utilization of
768 specified resource. Utilization option should be in format
769 name=value, value has to be integer. Options may be removed by
770 setting an option without a value. Example: pcs resource uti‐
771 lization TestResource cpu= ram=20 For the utilization configu‐
772 ration to be in effect, cluster property 'placement-strategy'
773 must be configured accordingly.
774 relations <resource id> [--full]
776 Display relations of a resource specified by its id with other
777 resources in a tree structure. Supported types of resource rela‐
778 tions are: ordering constraints, ordering set constraints, rela‐
779 tions defined by resource hierarchy (clones, groups, bundles).
780 If --full is used, more verbose output will be printed.
781 cluster
783 setup <cluster name> (<node name> [addr=<node address>]...)... [trans‐
784 port knet|udp|udpu [<transport options>] [link <link options>]... [com‐
785 pression <compression options>] [crypto <crypto options>]] [totem
786 <totem options>] [quorum <quorum options>] [--no-cluster-uuid] ([--en‐
787 able] [--start [--wait[=<n>]]] [--no-keys-sync]) | [--corosync_conf
788 <path>]
789 Create a cluster from the listed nodes and synchronize cluster
790 configuration files to them. If --corosync_conf is specified, do
791 not connect to other nodes and save corosync.conf to the speci‐
792 fied path; see 'Local only mode' below for details.
793 Nodes are specified by their names and optionally their ad‐
795 dresses. If no addresses are specified for a node, pcs will con‐
796 figure corosync to communicate with that node using an address
797 provided in 'pcs host auth' command. Otherwise, pcs will config‐
798 ure corosync to communicate with the node using the specified
799 addresses.
800 Transport knet:
802 This is the default transport. It allows configuring traffic en‐
803 cryption and compression as well as using multiple addresses
804 (links) for nodes.
805 Transport options are: ip_version, knet_pmtud_interval,
806 link_mode
807 Link options are: link_priority, linknumber, mcastport, ping_in‐
808 terval, ping_precision, ping_timeout, pong_count, transport (udp
809 or sctp)
810 Each 'link' followed by options sets options for one link in the
811 order the links are defined by nodes' addresses. You can set
812 link options for a subset of links using a linknumber. See exam‐
813 ples below.
814 Compression options are: level, model, threshold
815 Crypto options are: cipher, hash, model
816 By default, encryption is enabled with cipher=aes256 and
817 hash=sha256. To disable encryption, set cipher=none and
818 hash=none.
819 Transports udp and udpu:
821 These transports are limited to one address per node. They do
822 not support traffic encryption nor compression.
823 Transport options are: ip_version, netmtu
824 Link options are: bindnetaddr, broadcast, mcastaddr, mcastport,
825 ttl
826 Totem and quorum can be configured regardless of used transport.
828 Totem options are: block_unlisted_ips, consensus, downcheck,
829 fail_recv_const, heartbeat_failures_allowed, hold, join,
830 max_messages, max_network_delay, merge, miss_count_const,
831 send_join, seqno_unchanged_const, token, token_coefficient, to‐
832 ken_retransmit, token_retransmits_before_loss_const, window_size
833 Quorum options are: auto_tie_breaker, last_man_standing,
834 last_man_standing_window, wait_for_all
835 Transports and their options, link, compression, crypto and
837 totem options are all documented in corosync.conf(5) man page;
838 knet link options are prefixed 'knet_' there, compression op‐
839 tions are prefixed 'knet_compression_' and crypto options are
840 prefixed 'crypto_'. Quorum options are documented in votequo‐
841 rum(5) man page.
842 --no-cluster-uuid will not generate a unique ID for the cluster.
844 --enable will configure the cluster to start on nodes boot.
845 --start will start the cluster right after creating it. --wait
846 will wait up to 'n' seconds for the cluster to start.
847 --no-keys-sync will skip creating and distributing pcsd SSL cer‐
848 tificate and key and corosync and pacemaker authkey files. Use
849 this if you provide your own certificates and keys.
850 Local only mode:
852 By default, pcs connects to all specified nodes to verify they
853 can be used in the new cluster and to send cluster configuration
854 files to them. If this is not what you want, specify
855 --corosync_conf option followed by a file path. Pcs will save
856 corosync.conf to the specified file and will not connect to
857 cluster nodes. These are the tasks that pcs skips in that case:
858 * make sure the nodes are not running or configured to run a
859 cluster already
860 * make sure cluster packages are installed on all nodes and
861 their versions are compatible
862 * make sure there are no cluster configuration files on any node
863 (run 'pcs cluster destroy' and remove pcs_settings.conf file on
864 all nodes)
865 * synchronize corosync and pacemaker authkeys, /etc/corosync/au‐
866 thkey and /etc/pacemaker/authkey respectively, and the
867 corosync.conf file
868 * authenticate the cluster nodes against each other ('pcs clus‐
869 ter auth' or 'pcs host auth' command)
870 * synchronize pcsd certificates (so that pcs web UI can be used
871 in an HA mode)
872 Examples:
874 Create a cluster with default settings:
875 pcs cluster setup newcluster node1 node2
876 Create a cluster using two links:
877 pcs cluster setup newcluster node1 addr=10.0.1.11
878 addr=10.0.2.11 node2 addr=10.0.1.12 addr=10.0.2.12
879 Set link options for all links. Link options are matched to the
880 links in order. The first link (link 0) has sctp transport, the
881 second link (link 1) has mcastport 55405:
882 pcs cluster setup newcluster node1 addr=10.0.1.11
883 addr=10.0.2.11 node2 addr=10.0.1.12 addr=10.0.2.12 transport
884 knet link transport=sctp link mcastport=55405
885 Set link options for the second and fourth links only. Link op‐
886 tions are matched to the links based on the linknumber option
887 (the first link is link 0):
888 pcs cluster setup newcluster node1 addr=10.0.1.11
889 addr=10.0.2.11 addr=10.0.3.11 addr=10.0.4.11 node2
890 addr=10.0.1.12 addr=10.0.2.12 addr=10.0.3.12 addr=10.0.4.12
891 transport knet link linknumber=3 mcastport=55405 link linknum‐
892 ber=1 transport=sctp
893 Create a cluster using udp transport with a non-default port:
894 pcs cluster setup newcluster node1 node2 transport udp link
895 mcastport=55405
896 config [show] [--output-format text|cmd|json] [--corosync_conf <path>]
898 Show cluster configuration. There are 3 formats of output avail‐
899 able: 'cmd', 'json' and 'text', default is 'text'. Format 'text'
900 is a human friendly output. Format 'cmd' prints pcs commands
901 which can be used to recreate the same configuration. Format
902 'json' is a machine oriented output of the configuration. If
903 --corosync_conf is specified, configuration file specified by
904 <path> is used instead of the current cluster configuration.
905 config update [transport <transport options>] [compression <compression
907 options>] [crypto <crypto options>] [totem <totem options>]
908 [--corosync_conf <path>]
909 Update cluster configuration. Unspecified options will be kept
910 unchanged. If you wish to remove an option, set it to empty
911 value, i.e. 'option_name='.
912 If --corosync_conf is specified, update cluster configuration in
914 a file specified by <path>.
915 All options are documented in corosync.conf(5) man page. There
917 are different transport options for transport types. Compression
918 and crypto options are only available for knet transport. Totem
919 options can be set regardless of the transport type.
920 Transport options for knet transport are: ip_version, knet_pm‐
921 tud_interval, link_mode
922 Transport options for udp and updu transports are: ip_version,
923 netmtu
924 Compression options are: level, model, threshold
925 Crypto options are: cipher, hash, model
926 Totem options are: block_unlisted_ips, consensus, downcheck,
927 fail_recv_const, heartbeat_failures_allowed, hold, join,
928 max_messages, max_network_delay, merge, miss_count_const,
929 send_join, seqno_unchanged_const, token, token_coefficient, to‐
930 ken_retransmit, token_retransmits_before_loss_const, window_size
931 config uuid generate [--corosync_conf <path>] [--force]
933 Generate a new cluster UUID and distribute it to all cluster
934 nodes. Cluster UUID is not used by the cluster stack in any way,
935 it is provided to easily distinguish between multiple clusters
936 in a multi-cluster environment since the cluster name does not
937 have to be unique.
938 If --corosync_conf is specified, update cluster configuration in
940 file specified by <path>.
941 If --force is specified, existing UUID will be overwritten.
943 authkey corosync [<path>]
945 Generate a new corosync authkey and distribute it to all cluster
946 nodes. If <path> is specified, do not generate a key and use key
947 from the file.
948 start [--all | <node>... ] [--wait[=<n>]] [--request-timeout=<seconds>]
950 Start a cluster on specified node(s). If no nodes are specified
951 then start a cluster on the local node. If --all is specified
952 then start a cluster on all nodes. If the cluster has many nodes
953 then the start request may time out. In that case you should
954 consider setting --request-timeout to a suitable value. If
955 --wait is specified, pcs waits up to 'n' seconds for the cluster
956 to get ready to provide services after the cluster has success‐
957 fully started.
958 stop [--all | <node>... ] [--request-timeout=<seconds>]
960 Stop a cluster on specified node(s). If no nodes are specified
961 then stop a cluster on the local node. If --all is specified
962 then stop a cluster on all nodes. If the cluster is running re‐
963 sources which take long time to stop then the stop request may
964 time out before the cluster actually stops. In that case you
965 should consider setting --request-timeout to a suitable value.
966 kill Force corosync and pacemaker daemons to stop on the local node
968 (performs kill -9). Note that init system (e.g. systemd) can de‐
969 tect that cluster is not running and start it again. If you want
970 to stop cluster on a node, run pcs cluster stop on that node.
971 enable [--all | <node>... ]
973 Configure cluster to run on node boot on specified node(s). If
974 node is not specified then cluster is enabled on the local node.
975 If --all is specified then cluster is enabled on all nodes.
976 disable [--all | <node>... ]
978 Configure cluster to not run on node boot on specified node(s).
979 If node is not specified then cluster is disabled on the local
980 node. If --all is specified then cluster is disabled on all
981 nodes.
982 auth [-u <username>] [-p <password>]
984 Authenticate pcs/pcsd to pcsd on nodes configured in the local
985 cluster.
986 status View current cluster status (an alias of 'pcs status cluster').
988 sync Sync cluster configuration (files which are supported by all
990 subcommands of this command) to all cluster nodes.
991 sync corosync
993 Sync corosync configuration to all nodes found from current
994 corosync.conf file.
995 cib [filename] [scope=<scope> | --config]
997 Get the raw xml from the CIB (Cluster Information Base). If a
998 filename is provided, we save the CIB to that file, otherwise
999 the CIB is printed. Specify scope to get a specific section of
1000 the CIB. Valid values of the scope are: acls, alerts, configura‐
1001 tion, constraints, crm_config, fencing-topology, nodes, op_de‐
1002 faults, resources, rsc_defaults, tags. --config is the same as
1003 scope=configuration. Do not specify a scope if you want to edit
1004 the saved CIB using pcs (pcs -f <command>).
1005 cib-push <filename> [--wait[=<n>]] [diff-against=<filename_original> |
1007 scope=<scope> | --config]
1008 Push the raw xml from <filename> to the CIB (Cluster Information
1009 Base). You can obtain the CIB by running the 'pcs cluster cib'
1010 command, which is recommended first step when you want to per‐
1011 form desired modifications (pcs -f <command>) for the one-off
1012 push.
1013 If diff-against is specified, pcs diffs contents of filename
1014 against contents of filename_original and pushes the result to
1015 the CIB.
1016 Specify scope to push a specific section of the CIB. Valid val‐
1017 ues of the scope are: acls, alerts, configuration, constraints,
1018 crm_config, fencing-topology, nodes, op_defaults, resources,
1019 rsc_defaults, tags. --config is the same as scope=configuration.
1020 Use of --config is recommended. Do not specify a scope if you
1021 need to push the whole CIB or be warned in the case of outdated
1022 CIB.
1023 If --wait is specified wait up to 'n' seconds for changes to be
1024 applied.
1025 WARNING: the selected scope of the CIB will be overwritten by
1026 the current content of the specified file.
1027 Example:
1029 pcs cluster cib > original.xml
1030 cp original.xml new.xml
1031 pcs -f new.xml constraint location apache prefers node2
1032 pcs cluster cib-push new.xml diff-against=original.xml
1033 cib-upgrade
1035 Upgrade the CIB to conform to the latest version of the document
1036 schema.
1037 edit [scope=<scope> | --config]
1039 Edit the cib in the editor specified by the $EDITOR environment
1040 variable and push out any changes upon saving. Specify scope to
1041 edit a specific section of the CIB. Valid values of the scope
1042 are: acls, alerts, configuration, constraints, crm_config, fenc‐
1043 ing-topology, nodes, op_defaults, resources, rsc_defaults, tags.
1044 --config is the same as scope=configuration. Use of --config is
1045 recommended. Do not specify a scope if you need to edit the
1046 whole CIB or be warned in the case of outdated CIB.
1047 node add <node name> [addr=<node address>]... [watchdog=<watchdog
1049 path>] [device=<SBD device path>]... [--start [--wait[=<n>]]] [--en‐
1050 able] [--no-watchdog-validation]
1051 Add the node to the cluster and synchronize all relevant config‐
1052 uration files to the new node. This command can only be run on
1053 an existing cluster node.
1054 The new node is specified by its name and optionally its ad‐
1056 dresses. If no addresses are specified for the node, pcs will
1057 configure corosync to communicate with the node using an address
1058 provided in 'pcs host auth' command. Otherwise, pcs will config‐
1059 ure corosync to communicate with the node using the specified
1060 addresses.
1061 Use 'watchdog' to specify a path to a watchdog on the new node,
1063 when SBD is enabled in the cluster. If SBD is configured with
1064 shared storage, use 'device' to specify path to shared device(s)
1065 on the new node.
1066 If --start is specified also start cluster on the new node, if
1068 --wait is specified wait up to 'n' seconds for the new node to
1069 start. If --enable is specified configure cluster to start on
1070 the new node on boot. If --no-watchdog-validation is specified,
1071 validation of watchdog will be skipped.
1072 WARNING: By default, it is tested whether the specified watchdog
1074 is supported. This may cause a restart of the system when a
1075 watchdog with no-way-out-feature enabled is present. Use
1076 --no-watchdog-validation to skip watchdog validation.
1077 node delete <node name> [<node name>]...
1079 Shutdown specified nodes and remove them from the cluster.
1080 node remove <node name> [<node name>]...
1082 Shutdown specified nodes and remove them from the cluster.
1083 node add-remote <node name> [<node address>] [options] [op <operation
1085 action> <operation options> [<operation action> <operation op‐
1086 tions>]...] [meta <meta options>] [--wait[=<n>]]
1087 Add the node to the cluster as a remote node. Sync all relevant
1088 configuration files to the new node. Start the node and config‐
1089 ure it to start the cluster on boot. Options are port and recon‐
1090 nect_interval. Operations and meta belong to an underlying con‐
1091 nection resource (ocf:pacemaker:remote). If node address is not
1092 specified for the node, pcs will configure pacemaker to communi‐
1093 cate with the node using an address provided in 'pcs host auth'
1094 command. Otherwise, pcs will configure pacemaker to communicate
1095 with the node using the specified addresses. If --wait is speci‐
1096 fied, wait up to 'n' seconds for the node to start.
1097 node delete-remote <node identifier>
1099 Shutdown specified remote node and remove it from the cluster.
1100 The node-identifier can be the name of the node or the address
1101 of the node.
1102 node remove-remote <node identifier>
1104 Shutdown specified remote node and remove it from the cluster.
1105 The node-identifier can be the name of the node or the address
1106 of the node.
1107 node add-guest <node name> <resource id> [options] [--wait[=<n>]]
1109 Make the specified resource a guest node resource. Sync all rel‐
1110 evant configuration files to the new node. Start the node and
1111 configure it to start the cluster on boot. Options are re‐
1112 mote-addr, remote-port and remote-connect-timeout. If re‐
1113 mote-addr is not specified for the node, pcs will configure
1114 pacemaker to communicate with the node using an address provided
1115 in 'pcs host auth' command. Otherwise, pcs will configure pace‐
1116 maker to communicate with the node using the specified ad‐
1117 dresses. If --wait is specified, wait up to 'n' seconds for the
1118 node to start.
1119 node delete-guest <node identifier>
1121 Shutdown specified guest node and remove it from the cluster.
1122 The node-identifier can be the name of the node or the address
1123 of the node or id of the resource that is used as the guest
1124 node.
1125 node remove-guest <node identifier>
1127 Shutdown specified guest node and remove it from the cluster.
1128 The node-identifier can be the name of the node or the address
1129 of the node or id of the resource that is used as the guest
1130 node.
1131 node clear <node name>
1133 Remove specified node from various cluster caches. Use this if a
1134 removed node is still considered by the cluster to be a member
1135 of the cluster.
1136 link add <node_name>=<node_address>... [options <link options>]
1138 Add a corosync link. One address must be specified for each
1139 cluster node. If no linknumber is specified, pcs will use the
1140 lowest available linknumber.
1141 Link options (documented in corosync.conf(5) man page) are:
1142 link_priority, linknumber, mcastport, ping_interval, ping_preci‐
1143 sion, ping_timeout, pong_count, transport (udp or sctp)
1144 link delete <linknumber> [<linknumber>]...
1146 Remove specified corosync links.
1147 link remove <linknumber> [<linknumber>]...
1149 Remove specified corosync links.
1150 link update <linknumber> [<node_name>=<node_address>...] [options <link
1152 options>]
1153 Add, remove or change node addresses / link options of an exist‐
1154 ing corosync link. Use this if you cannot add / remove links
1155 which is the preferred way. Unspecified options will be kept un‐
1156 changed. If you wish to remove an option, set it to empty value,
1157 i.e. 'option_name='.
1158 Link options (documented in corosync.conf(5) man page) are:
1159 for knet transport: link_priority, mcastport, ping_interval,
1160 ping_precision, ping_timeout, pong_count, transport (udp or
1161 sctp)
1162 for udp and udpu transports: bindnetaddr, broadcast, mcastaddr,
1163 mcastport, ttl
1164 uidgid List the current configured uids and gids of users allowed to
1166 connect to corosync.
1167 uidgid add [uid=<uid>] [gid=<gid>]
1169 Add the specified uid and/or gid to the list of users/groups al‐
1170 lowed to connect to corosync.
1171 uidgid delete [uid=<uid>] [gid=<gid>]
1173 Remove the specified uid and/or gid from the list of
1174 users/groups allowed to connect to corosync.
1175 uidgid remove [uid=<uid>] [gid=<gid>]
1177 Remove the specified uid and/or gid from the list of
1178 users/groups allowed to connect to corosync.
1179 corosync [node]
1181 Get the corosync.conf from the specified node or from the cur‐
1182 rent node if node not specified.
1183 reload corosync
1185 Reload the corosync configuration on the current node.
1186 destroy [--all] [--force]
1188 Permanently destroy the cluster on the current node, killing all
1189 cluster processes and removing all cluster configuration files.
1190 Using --all will attempt to destroy the cluster on all nodes in
1191 the local cluster.
1192 WARNING: This command permanently removes any cluster configura‐
1194 tion that has been created. It is recommended to run 'pcs clus‐
1195 ter stop' before destroying the cluster. To prevent accidental
1196 running of this command, --force or interactive user response is
1197 required in order to proceed.
1198 verify [--full] [-f <filename>]
1200 Checks the pacemaker configuration (CIB) for syntax and common
1201 conceptual errors. If no filename is specified the check is per‐
1202 formed on the currently running cluster. If --full is used more
1203 verbose output will be printed.
1204 report [--from "YYYY-M-D H:M:S" [--to "YYYY-M-D H:M:S"]] <dest>
1206 Create a tarball containing everything needed when reporting
1207 cluster problems. If --from and --to are not used, the report
1208 will include the past 24 hours.
1209 stonith
1211 [status [<resource id | tag id>] [node=<node>] [--hide-inactive]]
1212 Show status of all currently configured stonith devices. If
1213 --hide-inactive is specified, only show active stonith devices.
1214 If a resource or tag id is specified, only show status of the
1215 specified resource or resources in the specified tag. If node is
1216 specified, only show status of resources configured for the
1217 specified node.
1218 config [--output-format text|cmd|json] [<stonith id>]...
1220 Show options of all currently configured stonith devices or if
1221 stonith device ids are specified show the options for the speci‐
1222 fied stonith device ids. There are 3 formats of output avail‐
1223 able: 'cmd', 'json' and 'text', default is 'text'. Format 'text'
1224 is a human friendly output. Format 'cmd' prints pcs commands
1225 which can be used to recreate the same configuration. Format
1226 'json' is a machine oriented output of the configuration.
1227 list [filter] [--nodesc]
1229 Show list of all available stonith agents (if filter is provided
1230 then only stonith agents matching the filter will be shown). If
1231 --nodesc is used then descriptions of stonith agents are not
1232 printed.
1233 describe <stonith agent> [--full]
1235 Show options for specified stonith agent. If --full is speci‐
1236 fied, all options including advanced and deprecated ones are
1237 shown.
1238 create <stonith id> <stonith device type> [stonith device options] [op
1240 <operation action> <operation options> [<operation action> <operation
1241 options>]...] [meta <meta options>] [--group <group id> [--before
1242 <stonith id> | --after <stonith id>]] [--disabled] [--agent-validation]
1243 [--wait[=n]]
1244 Create stonith device with specified type and options. If
1245 --group is specified the stonith device is added to the group
1246 named. You can use --before or --after to specify the position
1247 of the added stonith device relatively to some stonith device
1248 already existing in the group. If--disabled is specified the
1249 stonith device is not used. If --agent-validation is specified,
1250 stonith agent validate-all action will be used to validate
1251 stonith device options. If --wait is specified, pcs will wait up
1252 to 'n' seconds for the stonith device to start and then return 0
1253 if the stonith device is started, or 1 if the stonith device has
1254 not yet started. If 'n' is not specified it defaults to 60 min‐
1255 utes.
1256 Example: Create a device for nodes node1 and node2
1258 pcs stonith create MyFence fence_virt pcmk_host_list=node1,node2
1259 Example: Use port p1 for node n1 and ports p2 and p3 for node n2
1260 pcs stonith create MyFence fence_virt
1261 'pcmk_host_map=n1:p1;n2:p2,p3'
1262 update <stonith id> [stonith options] [op [<operation action> <opera‐
1264 tion options>]...] [meta <meta options>] [--agent-validation]
1265 [--wait[=n]]
1266 Add, remove or change options of specified stonith device. Un‐
1267 specified options will be kept unchanged. If you wish to remove
1268 an option, set it to empty value, i.e. 'option_name='.
1269 If an operation (op) is specified it will update the first found
1271 operation with the same action on the specified stonith device.
1272 If no operation with that action exists then a new operation
1273 will be created. (WARNING: all existing options on the updated
1274 operation will be reset if not specified.) If you want to create
1275 multiple monitor operations you should use the 'op add' & 'op
1276 remove' commands.
1277 If --agent-validation is specified, stonith agent validate-all
1279 action will be used to validate stonith device options.
1280 If --wait is specified, pcs will wait up to 'n' seconds for the
1282 changes to take effect and then return 0 if the changes have
1283 been processed or 1 otherwise. If 'n' is not specified it de‐
1284 faults to 60 minutes.
1285 update-scsi-devices <stonith id> (set <device-path> [<device-path>...])
1287 | (add <device-path> [<device-path>...] delete|remove <device-path>
1288 [<device-path>...] )
1289 Update scsi fencing devices without affecting other resources.
1290 You must specify either list of set devices or at least one de‐
1291 vice for add or delete/remove devices. Stonith resource must be
1292 running on one cluster node. Each device will be unfenced on
1293 each cluster node running cluster. Supported fence agents:
1294 fence_scsi, fence_mpath.
1295 delete <stonith id>
1297 Remove stonith id from configuration.
1298 remove <stonith id>
1300 Remove stonith id from configuration.
1301 op add <stonith id> <operation action> [operation properties]
1303 Add operation for specified stonith device.
1304 op delete <stonith id> <operation action> [<operation properties>...]
1306 Remove specified operation (note: you must specify the exact op‐
1307 eration properties to properly remove an existing operation).
1308 op delete <operation id>
1310 Remove the specified operation id.
1311 op remove <stonith id> <operation action> [<operation properties>...]
1313 Remove specified operation (note: you must specify the exact op‐
1314 eration properties to properly remove an existing operation).
1315 op remove <operation id>
1317 Remove the specified operation id.
1318 op defaults [config] [--all] [--full] [--no-check-expired]
1320 This command is an alias of 'resource op defaults [config]' com‐
1321 mand.
1322 List currently configured default values for operations. If
1324 --all is specified, also list expired sets of values. If --full
1325 is specified, also list ids. If --no-expire-check is specified,
1326 do not evaluate whether sets of values are expired.
1327 op defaults <name>=<value>...
1329 This command is an alias of 'resource op defaults' command.
1330 Set default values for operations.
1332 NOTE: Defaults do not apply to resources / stonith devices which
1333 override them with their own defined values.
1334 op defaults set create [<set options>] [meta [<name>=<value>]...] [rule
1336 [<expression>]]
1337 This command is an alias of 'resource op defaults set create'
1338 command.
1339 Create a new set of default values for resource / stonith device
1341 operations. You may specify a rule describing resources /
1342 stonith devices and / or operations to which the set applies.
1343 Set options are: id, score
1345 Expression looks like one of the following:
1347 op <operation name> [interval=<interval>]
1348 resource [<standard>]:[<provider>]:[<type>]
1349 defined|not_defined <node attribute>
1350 <node attribute> lt|gt|lte|gte|eq|ne [string|integer|num‐
1351 ber|version] <value>
1352 date gt|lt <date>
1353 date in_range [<date>] to <date>
1354 date in_range <date> to duration <duration options>
1355 date-spec <date-spec options>
1356 <expression> and|or <expression>
1357 (<expression>)
1358 You may specify all or any of 'standard', 'provider' and 'type'
1360 in a resource expression. For example: 'resource ocf::' matches
1361 all resources of 'ocf' standard, while 'resource ::Dummy'
1362 matches all resources of 'Dummy' type regardless of their stan‐
1363 dard and provider.
1364 Dates are expected to conform to ISO 8601 format.
1366 Duration options are: hours, monthdays, weekdays, yearsdays,
1368 months, weeks, years, weekyears, moon. Value for these options
1369 is an integer.
1370 Date-spec options are: hours, monthdays, weekdays, yearsdays,
1372 months, weeks, years, weekyears, moon. Value for these options
1373 is an integer or a range written as integer-integer.
1374 NOTE: Defaults do not apply to resources / stonith devices which
1376 override them with their own defined values.
1377 op defaults set delete [<set id>]...
1379 This command is an alias of 'resource op defaults set delete'
1380 command.
1381 Delete specified options sets.
1383 op defaults set remove [<set id>]...
1385 This command is an alias of 'resource op defaults set delete'
1386 command.
1387 Delete specified options sets.
1389 op defaults set update <set id> [meta [<name>=<value>]...]
1391 This command is an alias of 'resource op defaults set update'
1392 command.
1393 Add, remove or change values in specified set of default values
1395 for resource / stonith device operations. Unspecified options
1396 will be kept unchanged. If you wish to remove an option, set it
1397 to empty value, i.e. 'option_name='.
1398 NOTE: Defaults do not apply to resources / stonith devices which
1400 override them with their own defined values.
1401 op defaults update <name>=<value>...
1403 This command is an alias of 'resource op defaults update' com‐
1404 mand.
1405 Add, remove or change default values for operations. This is a
1407 simplified command useful for cases when you only manage one set
1408 of default values. Unspecified options will be kept unchanged.
1409 If you wish to remove an option, set it to empty value, i.e.
1410 'option_name='.
1411 NOTE: Defaults do not apply to resources / stonith devices which
1413 override them with their own defined values.
1414 meta <stonith id> <meta options> [--wait[=n]]
1416 Add specified options to the specified stonith device. Meta op‐
1417 tions should be in the format of name=value, options may be re‐
1418 moved by setting an option without a value. If --wait is speci‐
1419 fied, pcs will wait up to 'n' seconds for the changes to take
1420 effect and then return 0 if the changes have been processed or 1
1421 otherwise. If 'n' is not specified it defaults to 60 minutes.
1422 Example: pcs stonith meta test_stonith failure-timeout=50 re‐
1424 source-stickiness=
1425 defaults [config] [--all] [--full] [--no-check-expired]
1427 This command is an alias of 'resource defaults [config]' com‐
1428 mand.
1429 List currently configured default values for resources / stonith
1431 devices. If --all is specified, also list expired sets of val‐
1432 ues. If --full is specified, also list ids. If --no-expire-check
1433 is specified, do not evaluate whether sets of values are ex‐
1434 pired.
1435 defaults <name>=<value>...
1437 This command is an alias of 'resource defaults' command.
1438 Set default values for resources / stonith devices.
1440 NOTE: Defaults do not apply to resources / stonith devices which
1441 override them with their own defined values.
1442 defaults set create [<set options>] [meta [<name>=<value>]...] [rule
1444 [<expression>]]
1445 This command is an alias of 'resource defaults set create' com‐
1446 mand.
1447 Create a new set of default values for resources / stonith de‐
1449 vices. You may specify a rule describing resources / stonith de‐
1450 vices to which the set applies.
1451 Set options are: id, score
1453 Expression looks like one of the following:
1455 resource [<standard>]:[<provider>]:[<type>]
1456 date gt|lt <date>
1457 date in_range [<date>] to <date>
1458 date in_range <date> to duration <duration options>
1459 date-spec <date-spec options>
1460 <expression> and|or <expression>
1461 (<expression>)
1462 You may specify all or any of 'standard', 'provider' and 'type'
1464 in a resource expression. For example: 'resource ocf::' matches
1465 all resources of 'ocf' standard, while 'resource ::Dummy'
1466 matches all resources of 'Dummy' type regardless of their stan‐
1467 dard and provider.
1468 Dates are expected to conform to ISO 8601 format.
1470 Duration options are: hours, monthdays, weekdays, yearsdays,
1472 months, weeks, years, weekyears, moon. Value for these options
1473 is an integer.
1474 Date-spec options are: hours, monthdays, weekdays, yearsdays,
1476 months, weeks, years, weekyears, moon. Value for these options
1477 is an integer or a range written as integer-integer.
1478 NOTE: Defaults do not apply to resources / stonith devices which
1480 override them with their own defined values.
1481 defaults set delete [<set id>]...
1483 This command is an alias of 'resource defaults set delete' com‐
1484 mand.
1485 Delete specified options sets.
1487 defaults set remove [<set id>]...
1489 This command is an alias of 'resource defaults set delete' com‐
1490 mand.
1491 Delete specified options sets.
1493 defaults set update <set id> [meta [<name>=<value>]...]
1495 This command is an alias of 'resource defaults set update' com‐
1496 mand.
1497 Add, remove or change values in specified set of default values
1499 for resources / stonith devices. Unspecified options will be
1500 kept unchanged. If you wish to remove an option, set it to empty
1501 value, i.e. 'option_name='.
1502 NOTE: Defaults do not apply to resources / stonith devices which
1504 override them with their own defined values.
1505 defaults update <name>=<value>...
1507 This command is an alias of 'resource defaults update' command.
1508 Add, remove or change default values for resources / stonith de‐
1510 vices. This is a simplified command useful for cases when you
1511 only manage one set of default values. Unspecified options will
1512 be kept unchanged. If you wish to remove an option, set it to
1513 empty value, i.e. 'option_name='.
1514 NOTE: Defaults do not apply to resources / stonith devices which
1516 override them with their own defined values.
1517 cleanup [<resource id | stonith id>] [node=<node>] [operation=<opera‐
1519 tion> [interval=<interval>]] [--strict]
1520 This command is an alias of 'resource cleanup' command.
1521 Make the cluster forget failed operations from history of the
1523 resource / stonith device and re-detect its current state. This
1524 can be useful to purge knowledge of past failures that have
1525 since been resolved.
1526 If the named resource is part of a group, or one numbered in‐
1528 stance of a clone or bundled resource, the clean-up applies to
1529 the whole collective resource unless --strict is given.
1530 If a resource id / stonith id is not specified then all re‐
1532 sources / stonith devices will be cleaned up.
1533 If a node is not specified then resources / stonith devices on
1535 all nodes will be cleaned up.
1536 refresh [<resource id | stonith id>] [node=<node>] [--strict]
1538 This command is an alias of 'resource refresh' command.
1539 Make the cluster forget the complete operation history (includ‐
1541 ing failures) of the resource / stonith device and re-detect its
1542 current state. If you are interested in forgetting failed opera‐
1543 tions only, use the 'pcs resource cleanup' command.
1544 If the named resource is part of a group, or one numbered in‐
1546 stance of a clone or bundled resource, the refresh applies to
1547 the whole collective resource unless --strict is given.
1548 If a resource id / stonith id is not specified then all re‐
1550 sources / stonith devices will be refreshed.
1551 If a node is not specified then resources / stonith devices on
1553 all nodes will be refreshed.
1554 failcount [show [<resource id | stonith id>] [node=<node>] [opera‐
1556 tion=<operation> [interval=<interval>]]] [--full]
1557 This command is an alias of 'resource failcount show' command.
1558 Show current failcount for resources and stonith devices, op‐
1560 tionally filtered by a resource / stonith device, node, opera‐
1561 tion and its interval. If --full is specified do not sum fail‐
1562 counts per resource / stonith device and node. Use 'pcs resource
1563 cleanup' or 'pcs resource refresh' to reset failcounts.
1564 enable <stonith id>... [--wait[=n]]
1566 Allow the cluster to use the stonith devices. If --wait is spec‐
1567 ified, pcs will wait up to 'n' seconds for the stonith devices
1568 to start and then return 0 if the stonith devices are started,
1569 or 1 if the stonith devices have not yet started. If 'n' is not
1570 specified it defaults to 60 minutes.
1571 disable <stonith id>... [--wait[=n]]
1573 Attempt to stop the stonith devices if they are running and dis‐
1574 allow the cluster to use them. If --wait is specified, pcs will
1575 wait up to 'n' seconds for the stonith devices to stop and then
1576 return 0 if the stonith devices are stopped or 1 if the stonith
1577 devices have not stopped. If 'n' is not specified it defaults to
1578 60 minutes.
1579 level [config]
1581 Lists all of the fencing levels currently configured.
1582 level add <level> <target> <stonith id> [stonith id]...
1584 Add the fencing level for the specified target with the list of
1585 stonith devices to attempt for that target at that level. Fence
1586 levels are attempted in numerical order (starting with 1). If a
1587 level succeeds (meaning all devices are successfully fenced in
1588 that level) then no other levels are tried, and the target is
1589 considered fenced. Target may be a node name <node_name> or
1590 %<node_name> or node%<node_name>, a node name regular expression
1591 regexp%<node_pattern> or a node attribute value at‐
1592 trib%<name>=<value>.
1593 level delete <level> [target <target>] [stonith <stonith id>...]
1595 Removes the fence level for the level, target and/or devices
1596 specified. If no target or devices are specified then the fence
1597 level is removed. Target may be a node name <node_name> or
1598 %<node_name> or node%<node_name>, a node name regular expression
1599 regexp%<node_pattern> or a node attribute value at‐
1600 trib%<name>=<value>.
1601 level remove <level> [target <target>] [stonith <stonith id>...]
1603 Removes the fence level for the level, target and/or devices
1604 specified. If no target or devices are specified then the fence
1605 level is removed. Target may be a node name <node_name> or
1606 %<node_name> or node%<node_name>, a node name regular expression
1607 regexp%<node_pattern> or a node attribute value at‐
1608 trib%<name>=<value>.
1609 level clear [target <target> | stonith <stonith id>...]
1611 Clears the fence levels on the target (or stonith id) specified
1612 or clears all fence levels if a target/stonith id is not speci‐
1613 fied. Target may be a node name <node_name> or %<node_name> or
1614 node%<node_name>, a node name regular expression reg‐
1615 exp%<node_pattern> or a node attribute value at‐
1616 trib%<name>=<value>. Example: pcs stonith level clear stonith
1617 dev_a dev_b
1618 level verify
1620 Verifies all fence devices and nodes specified in fence levels
1621 exist.
1622 fence <node> [--off]
1624 Fence the node specified (if --off is specified, use the 'off'
1625 API call to stonith which will turn the node off instead of re‐
1626 booting it).
1627 confirm <node> [--force]
1629 Confirm to the cluster that the specified node is powered off.
1630 This allows the cluster to recover from a situation where no
1631 stonith device is able to fence the node. This command should
1632 ONLY be used after manually ensuring that the node is powered
1633 off and has no access to shared resources.
1634 WARNING: If this node is not actually powered off or it does
1636 have access to shared resources, data corruption/cluster failure
1637 can occur. To prevent accidental running of this command,
1638 --force or interactive user response is required in order to
1639 proceed.
1640 NOTE: It is not checked if the specified node exists in the
1642 cluster in order to be able to work with nodes not visible from
1643 the local cluster partition.
1644 history [show [<node>]]
1646 Show fencing history for the specified node or all nodes if no
1647 node specified.
1648 history cleanup [<node>]
1650 Cleanup fence history of the specified node or all nodes if no
1651 node specified.
1652 history update
1654 Update fence history from all nodes.
1655 sbd enable [watchdog=<path>[@<node>]]... [device=<path>[@<node>]]...
1657 [<SBD_OPTION>=<value>]... [--no-watchdog-validation]
1658 Enable SBD in cluster. Default path for watchdog device is
1659 /dev/watchdog. Allowed SBD options: SBD_WATCHDOG_TIMEOUT (de‐
1660 fault: 5), SBD_DELAY_START (default: no), SBD_STARTMODE (de‐
1661 fault: always) and SBD_TIMEOUT_ACTION. SBD options are docu‐
1662 mented in sbd(8) man page. It is possible to specify up to 3 de‐
1663 vices per node. If --no-watchdog-validation is specified, vali‐
1664 dation of watchdogs will be skipped.
1665 WARNING: Cluster has to be restarted in order to apply these
1667 changes.
1668 WARNING: By default, it is tested whether the specified watchdog
1670 is supported. This may cause a restart of the system when a
1671 watchdog with no-way-out-feature enabled is present. Use
1672 --no-watchdog-validation to skip watchdog validation.
1673 Example of enabling SBD in cluster with watchdogs on node1 will
1675 be /dev/watchdog2, on node2 /dev/watchdog1, /dev/watchdog0 on
1676 all other nodes, device /dev/sdb on node1, device /dev/sda on
1677 all other nodes and watchdog timeout will bet set to 10 seconds:
1678 pcs stonith sbd enable watchdog=/dev/watchdog2@node1 watch‐
1680 dog=/dev/watchdog1@node2 watchdog=/dev/watchdog0 de‐
1681 vice=/dev/sdb@node1 device=/dev/sda SBD_WATCHDOG_TIMEOUT=10
1682 sbd disable
1685 Disable SBD in cluster.
1686 WARNING: Cluster has to be restarted in order to apply these
1688 changes.
1689 sbd device setup device=<path> [device=<path>]... [watchdog-time‐
1691 out=<integer>] [allocate-timeout=<integer>] [loop-timeout=<integer>]
1692 [msgwait-timeout=<integer>]
1693 Initialize SBD structures on device(s) with specified timeouts.
1694 WARNING: All content on device(s) will be overwritten.
1696 sbd device message <device-path> <node> <message-type>
1698 Manually set a message of the specified type on the device for
1699 the node. Possible message types (they are documented in sbd(8)
1700 man page): test, reset, off, crashdump, exit, clear
1701 sbd status [--full]
1703 Show status of SBD services in cluster and local device(s) con‐
1704 figured. If --full is specified, also dump of SBD headers on de‐
1705 vice(s) will be shown.
1706 sbd config
1708 Show SBD configuration in cluster.
1709 sbd watchdog list
1712 Show all available watchdog devices on the local node.
1713 WARNING: Listing available watchdogs may cause a restart of the
1715 system when a watchdog with no-way-out-feature enabled is
1716 present.
1717 sbd watchdog test [<watchdog-path>]
1720 This operation is expected to force-reboot the local system
1721 without following any shutdown procedures using a watchdog. If
1722 no watchdog is specified, available watchdog will be used if
1723 only one watchdog device is available on the local system.
1724 acl
1727 [config]
1728 List all current access control lists.
1729 enable Enable access control lists.
1731 disable
1733 Disable access control lists.
1734 role create <role id> [description=<description>] [((read | write |
1736 deny) (xpath <query> | id <id>))...]
1737 Create a role with the id and (optional) description specified.
1738 Each role can also have an unlimited number of permissions
1739 (read/write/deny) applied to either an xpath query or the id of
1740 a specific element in the cib.
1741 Permissions are applied to the selected XML element's entire XML
1742 subtree (all elements enclosed within it). Write permission
1743 grants the ability to create, modify, or remove the element and
1744 its subtree, and also the ability to create any "scaffolding"
1745 elements (enclosing elements that do not have attributes other
1746 than an ID). Permissions for more specific matches (more deeply
1747 nested elements) take precedence over more general ones. If mul‐
1748 tiple permissions are configured for the same match (for exam‐
1749 ple, in different roles applied to the same user), any deny per‐
1750 mission takes precedence, then write, then lastly read.
1751 An xpath may include an attribute expression to select only ele‐
1752 ments that match the expression, but the permission still ap‐
1753 plies to the entire element (and its subtree), not to the attri‐
1754 bute alone. For example, using the xpath "//*[@name]" to give
1755 write permission would allow changes to the entirety of all ele‐
1756 ments that have a "name" attribute and everything enclosed by
1757 those elements. There is no way currently to give permissions
1758 for just one attribute of an element. That is to say, you can
1759 not define an ACL that allows someone to read just the dc-uuid
1760 attribute of the cib tag - that would select the cib element and
1761 give read access to the entire CIB.
1762 role delete <role id>
1764 Delete the role specified and remove it from any users/groups it
1765 was assigned to.
1766 role remove <role id>
1768 Delete the role specified and remove it from any users/groups it
1769 was assigned to.
1770 role assign <role id> [to] [user|group] <username/group>
1772 Assign a role to a user or group already created with 'pcs acl
1773 user/group create'. If there is user and group with the same id
1774 and it is not specified which should be used, user will be pri‐
1775 oritized. In cases like this specify whenever user or group
1776 should be used.
1777 role unassign <role id> [from] [user|group] <username/group>
1779 Remove a role from the specified user. If there is user and
1780 group with the same id and it is not specified which should be
1781 used, user will be prioritized. In cases like this specify when‐
1782 ever user or group should be used.
1783 user create <username> [<role id>]...
1785 Create an ACL for the user specified and assign roles to the
1786 user.
1787 user delete <username>
1789 Remove the user specified (and roles assigned will be unassigned
1790 for the specified user).
1791 user remove <username>
1793 Remove the user specified (and roles assigned will be unassigned
1794 for the specified user).
1795 group create <group> [<role id>]...
1797 Create an ACL for the group specified and assign roles to the
1798 group.
1799 group delete <group>
1801 Remove the group specified (and roles assigned will be unas‐
1802 signed for the specified group).
1803 group remove <group>
1805 Remove the group specified (and roles assigned will be unas‐
1806 signed for the specified group).
1807 permission add <role id> ((read | write | deny) (xpath <query> | id
1809 <id>))...
1810 Add the listed permissions to the role specified. Permissions
1811 are applied to either an xpath query or the id of a specific el‐
1812 ement in the CIB.
1813 Permissions are applied to the selected XML element's entire XML
1814 subtree (all elements enclosed within it). Write permission
1815 grants the ability to create, modify, or remove the element and
1816 its subtree, and also the ability to create any "scaffolding"
1817 elements (enclosing elements that do not have attributes other
1818 than an ID). Permissions for more specific matches (more deeply
1819 nested elements) take precedence over more general ones. If mul‐
1820 tiple permissions are configured for the same match (for exam‐
1821 ple, in different roles applied to the same user), any deny per‐
1822 mission takes precedence, then write, then lastly read.
1823 An xpath may include an attribute expression to select only ele‐
1824 ments that match the expression, but the permission still ap‐
1825 plies to the entire element (and its subtree), not to the attri‐
1826 bute alone. For example, using the xpath "//*[@name]" to give
1827 write permission would allow changes to the entirety of all ele‐
1828 ments that have a "name" attribute and everything enclosed by
1829 those elements. There is no way currently to give permissions
1830 for just one attribute of an element. That is to say, you can
1831 not define an ACL that allows someone to read just the dc-uuid
1832 attribute of the cib tag - that would select the cib element and
1833 give read access to the entire CIB.
1834 permission delete <permission id>
1836 Remove the permission id specified (permission id's are listed
1837 in parenthesis after permissions in 'pcs acl' output).
1838 permission remove <permission id>
1840 Remove the permission id specified (permission id's are listed
1841 in parenthesis after permissions in 'pcs acl' output).
1842 property
1844 [config [<property>... | --all | --defaults | (--output-format
1845 text|cmd|json)]] | [--all | --defaults | (--output-format
1846 text|cmd|json)]
1847 List property settings (default: lists configured properties).
1848 If --defaults is specified will show all property defaults, if
1849 --all is specified, current configured properties will be shown
1850 with unset properties and their defaults. See 'pcs property de‐
1851 scribe' for a description of the properties. There are 3 formats
1852 of output available: 'cmd', 'json' and 'text', default is
1853 'text'. Format 'text' is a human friendly output. Format 'cmd'
1854 prints pcs commands which can be used to recreate the same con‐
1855 figuration. Format 'json' is a machine oriented output of the
1856 configuration.
1857 defaults [<property>...] | [--full]
1859 List all property defaults or only defaults for specified prop‐
1860 erties. If --full is specified, all properties defaults includ‐
1861 ing advanced are shown.
1862 describe [<property>...] [--output-format text|json] [--full]
1865 Show cluster properties. There are 2 formats of output avail‐
1866 able: 'json' and 'text', default is 'text'. Format 'text' is a
1867 human friendly output. Format 'json' is a machine oriented out‐
1868 put of the configuration. If --full is specified, all properties
1869 descriptions including advanced are shown.
1870 set <property>=[<value>] ... [--force]
1872 Set specific pacemaker properties (if the value is blank then
1873 the property is removed from the configuration). If a property
1874 is not recognized by pcs the property will not be created unless
1875 the --force is used. See 'pcs property describe' for a descrip‐
1876 tion of the properties.
1877 unset <property> ...
1879 Remove property from configuration. See 'pcs property describe'
1880 for a description of the properties.
1881 constraint
1883 [config] [--full] [--all] [--output-format text|cmd|json]
1884 List all current constraints that are not expired. If --all is
1885 specified also show expired constraints. If --full is specified
1886 also list the constraint ids. There are 3 formats of output
1887 available: 'cmd', 'json' and 'text', default is 'text'. Format
1888 'text' is a human friendly output. Format 'cmd' prints pcs com‐
1889 mands which can be used to recreate the same configuration. For‐
1890 mat 'json' is a machine oriented output of the configuration.
1891 location <resource> prefers <node>[=<score>] [<node>[=<score>]]...
1893 Create a location constraint on a resource to prefer the speci‐
1894 fied node with score (default score: INFINITY). Resource may be
1895 either a resource id <resource_id> or %<resource_id> or re‐
1896 source%<resource_id>, or a resource name regular expression reg‐
1897 exp%<resource_pattern>.
1898 location <resource> avoids <node>[=<score>] [<node>[=<score>]]...
1900 Create a location constraint on a resource to avoid the speci‐
1901 fied node with score (default score: INFINITY). Resource may be
1902 either a resource id <resource_id> or %<resource_id> or re‐
1903 source%<resource_id>, or a resource name regular expression reg‐
1904 exp%<resource_pattern>.
1905 location <resource> rule [id=<rule id>] [resource-discovery=<option>]
1907 [role=Promoted|Unpromoted] [constraint-id=<id>] [score=<score> |
1908 score-attribute=<attribute>] <expression>
1909 Creates a location constraint with a rule on the specified re‐
1910 source where expression looks like one of the following:
1911 defined|not_defined <node attribute>
1912 <node attribute> lt|gt|lte|gte|eq|ne [string|integer|num‐
1913 ber|version] <value>
1914 date gt|lt <date>
1915 date in_range <date> to <date>
1916 date in_range <date> to duration <duration options>...
1917 date-spec <date spec options>...
1918 <expression> and|or <expression>
1919 ( <expression> )
1920 where duration options and date spec options are: hours, month‐
1921 days, weekdays, yeardays, months, weeks, years, weekyears, moon.
1922 Resource may be either a resource id <resource_id> or %<re‐
1923 source_id> or resource%<resource_id>, or a resource name regular
1924 expression regexp%<resource_pattern>. If score is omitted it de‐
1925 faults to INFINITY. If id is omitted one is generated from the
1926 resource id. If resource-discovery is omitted it defaults to
1927 'always'.
1928 location [config [resources [<resource reference>...]] | [nodes
1930 [<node>...]]] [--full] [--all] [--output-format text|cmd|json]
1931 List all the current location constraints that are not expired.
1932 If 'resources' is specified, location constraints are displayed
1933 per resource. If 'nodes' is specified, location constraints are
1934 displayed per node. If specific nodes, resources or resource
1935 name regular expressions are specified, only constraints con‐
1936 taining those will be shown. Resource reference may be either a
1937 resource id <resource_id> or %<resource_id> or resource%<re‐
1938 source_id>, or a resource name regular expression regexp%<re‐
1939 source_pattern>. If --full is specified show the internal con‐
1940 straint id's as well. If --all is specified show the expired
1941 constraints. There are 3 formats of output available: 'cmd',
1942 'json' and 'text', default is 'text'. Format 'text' is a human
1943 friendly output. Format 'cmd' prints pcs commands which can be
1944 used to recreate the same configuration. Format 'json' is a ma‐
1945 chine oriented output of the configuration.
1946 location add <id> <resource> <node> <score> [resource-discovery=<op‐
1948 tion>]
1949 Add a location constraint with the appropriate id for the speci‐
1950 fied resource, node name and score. Resource may be either a re‐
1951 source id <resource_id> or %<resource_id> or resource%<re‐
1952 source_id>, or a resource name regular expression regexp%<re‐
1953 source_pattern>.
1954 location delete <id>
1956 Remove a location constraint with the appropriate id.
1957 location remove <id>
1959 Remove a location constraint with the appropriate id.
1960 order [config] [--full] [--output-format text|cmd|json]
1962 List all current ordering constraints (if --full is specified
1963 show the internal constraint id's as well). There are 3 formats
1964 of output available: 'cmd', 'json' and 'text', default is
1965 'text'. Format 'text' is a human friendly output. Format 'cmd'
1966 prints pcs commands which can be used to recreate the same con‐
1967 figuration. Format 'json' is a machine oriented output of the
1968 configuration.
1969 order [action] <resource id> then [action] <resource id> [options]
1971 Add an ordering constraint specifying actions (start, stop, pro‐
1972 mote, demote) and if no action is specified the default action
1973 will be start. Available options are kind=Optional/Manda‐
1974 tory/Serialize, symmetrical=true/false, require-all=true/false
1975 and id=<constraint-id>.
1976 order set <resource1> [resourceN]... [options] [set <resourceX> ...
1978 [options]] [setoptions [constraint_options]]
1979 Create an ordered set of resources. Available options are se‐
1980 quential=true/false, require-all=true/false and ac‐
1981 tion=start/promote/demote/stop. Available constraint_options are
1982 id=<constraint-id>, kind=Optional/Mandatory/Serialize and sym‐
1983 metrical=true/false.
1984 order delete <resource1> [resourceN]...
1986 Remove resource from any ordering constraint
1987 order remove <resource1> [resourceN]...
1989 Remove resource from any ordering constraint
1990 colocation [config] [--full] [--output-format text|cmd|json]
1992 List all current colocation constraints (if --full is specified
1993 show the internal constraint id's as well). There are 3 formats
1994 of output available: 'cmd', 'json' and 'text', default is
1995 'text'. Format 'text' is a human friendly output. Format 'cmd'
1996 prints pcs commands which can be used to recreate the same con‐
1997 figuration. Format 'json' is a machine oriented output of the
1998 configuration.
1999 colocation add [<role>] <source resource id> with [<role>] <target re‐
2001 source id> [score] [options] [id=constraint-id]
2002 Request <source resource> to run on the same node where pace‐
2003 maker has determined <target resource> should run. Positive
2004 values of score mean the resources should be run on the same
2005 node, negative values mean the resources should not be run on
2006 the same node. Specifying 'INFINITY' (or '-INFINITY') for the
2007 score forces <source resource> to run (or not run) with <target
2008 resource> (score defaults to "INFINITY"). A role can be: 'Pro‐
2009 moted', 'Unpromoted', 'Started', 'Stopped' (if no role is speci‐
2010 fied, it defaults to 'Started').
2011 colocation set <resource1> [resourceN]... [options] [set <resourceX>
2013 ... [options]] [setoptions [constraint_options]]
2014 Create a colocation constraint with a resource set. Available
2015 options are sequential=true/false and role=Stopped/Started/Pro‐
2016 moted/Unpromoted. Available constraint_options are id and score.
2017 colocation delete <source resource id> <target resource id>
2019 Remove colocation constraints with specified resources.
2020 colocation remove <source resource id> <target resource id>
2022 Remove colocation constraints with specified resources.
2023 ticket [config] [--full] [--output-format text|cmd|json]
2025 List all current ticket constraints (if --full is specified show
2026 the internal constraint id's as well). There are 3 formats of
2027 output available: 'cmd', 'json' and 'text', default is 'text'.
2028 Format 'text' is a human friendly output. Format 'cmd' prints
2029 pcs commands which can be used to recreate the same configura‐
2030 tion. Format 'json' is a machine oriented output of the configu‐
2031 ration.
2032 ticket add <ticket> [<role>] <resource id> [<options>] [id=<con‐
2034 straint-id>]
2035 Create a ticket constraint for <resource id>. Available option
2036 is loss-policy=fence/stop/freeze/demote. A role can be Promoted,
2037 Unpromoted, Started or Stopped.
2038 ticket set <resource1> [<resourceN>]... [<options>] [set <resourceX>
2040 ... [<options>]] setoptions <constraint_options>
2041 Create a ticket constraint with a resource set. Available op‐
2042 tions are role=Stopped/Started/Promoted/Unpromoted. Required
2043 constraint option is ticket=<ticket>. Optional constraint op‐
2044 tions are id=<constraint-id> and loss-pol‐
2045 icy=fence/stop/freeze/demote.
2046 ticket delete <ticket> <resource id>
2048 Remove all ticket constraints with <ticket> from <resource id>.
2049 ticket remove <ticket> <resource id>
2051 Remove all ticket constraints with <ticket> from <resource id>.
2052 delete <constraint id>...
2054 Remove constraint(s) or constraint rules with the specified
2055 id(s).
2056 remove <constraint id>...
2058 Remove constraint(s) or constraint rules with the specified
2059 id(s).
2060 ref <resource>...
2062 List constraints referencing specified resource.
2063 rule add <constraint id> [id=<rule id>] [role=Promoted|Unpromoted]
2065 [score=<score>|score-attribute=<attribute>] <expression>
2066 Add a rule to a location constraint specified by 'constraint id'
2067 where the expression looks like one of the following:
2068 defined|not_defined <node attribute>
2069 <node attribute> lt|gt|lte|gte|eq|ne [string|integer|num‐
2070 ber|version] <value>
2071 date gt|lt <date>
2072 date in_range <date> to <date>
2073 date in_range <date> to duration <duration options>...
2074 date-spec <date spec options>...
2075 <expression> and|or <expression>
2076 ( <expression> )
2077 where duration options and date spec options are: hours, month‐
2078 days, weekdays, yeardays, months, weeks, years, weekyears, moon.
2079 If score is omitted it defaults to INFINITY. If id is omitted
2080 one is generated from the constraint id.
2081 rule delete <rule id>
2083 Remove a rule from its location constraint and if it's the last
2084 rule, the constraint will also be removed.
2085 rule remove <rule id>
2087 Remove a rule from its location constraint and if it's the last
2088 rule, the constraint will also be removed.
2089 qdevice
2091 status <device model> [--full] [<cluster name>]
2092 Show runtime status of specified model of quorum device
2093 provider. Using --full will give more detailed output. If
2094 <cluster name> is specified, only information about the speci‐
2095 fied cluster will be displayed.
2096 setup model <device model> [--enable] [--start]
2098 Configure specified model of quorum device provider. Quorum de‐
2099 vice then can be added to clusters by running "pcs quorum device
2100 add" command in a cluster. --start will also start the
2101 provider. --enable will configure the provider to start on
2102 boot.
2103 destroy <device model>
2105 Disable and stop specified model of quorum device provider and
2106 delete its configuration files.
2107 start <device model>
2109 Start specified model of quorum device provider.
2110 stop <device model>
2112 Stop specified model of quorum device provider.
2113 kill <device model>
2115 Force specified model of quorum device provider to stop (per‐
2116 forms kill -9). Note that init system (e.g. systemd) can detect
2117 that the qdevice is not running and start it again. If you want
2118 to stop the qdevice, run "pcs qdevice stop" command.
2119 enable <device model>
2121 Configure specified model of quorum device provider to start on
2122 boot.
2123 disable <device model>
2125 Configure specified model of quorum device provider to not start
2126 on boot.
2127 quorum
2129 [config]
2130 Show quorum configuration.
2131 status Show quorum runtime status.
2133 device add [<generic options>] model <device model> [<model options>]
2135 [heuristics <heuristics options>]
2136 Add a quorum device to the cluster. Quorum device should be con‐
2137 figured first with "pcs qdevice setup". It is not possible to
2138 use more than one quorum device in a cluster simultaneously.
2139 Currently the only supported model is 'net'. It requires model
2140 options 'algorithm' and 'host' to be specified. Options are doc‐
2141 umented in corosync-qdevice(8) man page; generic options are
2142 'sync_timeout' and 'timeout', for model net options check the
2143 quorum.device.net section, for heuristics options see the quo‐
2144 rum.device.heuristics section. Pcs automatically creates and
2145 distributes TLS certificates and sets the 'tls' model option to
2146 the default value 'on'.
2147 Example: pcs quorum device add model net algorithm=lms
2148 host=qnetd.internal.example.com
2149 device heuristics delete
2151 Remove all heuristics settings of the configured quorum device.
2152 device heuristics remove
2154 Remove all heuristics settings of the configured quorum device.
2155 device delete
2157 Remove a quorum device from the cluster.
2158 device remove
2160 Remove a quorum device from the cluster.
2161 device status [--full]
2163 Show quorum device runtime status. Using --full will give more
2164 detailed output.
2165 device update [<generic options>] [model <model options>] [heuristics
2167 <heuristics options>]
2168 Add, remove or change quorum device options. Unspecified options
2169 will be kept unchanged. If you wish to remove an option, set it
2170 to empty value, i.e. 'option_name='. Requires the cluster to be
2171 stopped. Model and options are all documented in corosync-qde‐
2172 vice(8) man page; for heuristics options check the quorum.de‐
2173 vice.heuristics subkey section, for model options check the quo‐
2174 rum.device.<device model> subkey sections.
2175 WARNING: If you want to change "host" option of qdevice model
2177 net, use "pcs quorum device remove" and "pcs quorum device add"
2178 commands to set up configuration properly unless old and new
2179 host is the same machine.
2180 expected-votes <votes>
2182 Set expected votes in the live cluster to specified value. This
2183 only affects the live cluster, not changes any configuration
2184 files.
2185 unblock [--force]
2187 Cancel waiting for all nodes when establishing quorum. Useful
2188 in situations where you know the cluster is inquorate, but you
2189 are confident that the cluster should proceed with resource man‐
2190 agement regardless. This command should ONLY be used when nodes
2191 which the cluster is waiting for have been confirmed to be pow‐
2192 ered off and to have no access to shared resources.
2193 WARNING: If the nodes are not actually powered off or they do
2195 have access to shared resources, data corruption/cluster failure
2196 can occur. To prevent accidental running of this command,
2197 --force or interactive user response is required in order to
2198 proceed.
2199 update [auto_tie_breaker=[0|1]] [last_man_standing=[0|1]]
2201 [last_man_standing_window=[<time in ms>]] [wait_for_all=[0|1]]
2202 Add, remove or change quorum options. At least one option must
2203 be specified. Unspecified options will be kept unchanged. If you
2204 wish to remove an option, set it to empty value, i.e. 'op‐
2205 tion_name='. Options are documented in corosync's votequorum(5)
2206 man page. Requires the cluster to be stopped.
2207 booth
2209 setup sites <address> <address> [<address>...] [arbitrators <address>
2210 ...] [--force]
2211 Write new booth configuration with specified sites and arbitra‐
2212 tors. Total number of peers (sites and arbitrators) must be
2213 odd. When the configuration file already exists, command fails
2214 unless --force is specified.
2215 destroy
2217 Remove booth configuration files.
2218 ticket add <ticket> [<name>=<value> ...]
2220 Add new ticket to the current configuration. Ticket options are
2221 specified in booth manpage.
2222 ticket delete <ticket>
2224 Remove the specified ticket from the current configuration.
2225 ticket remove <ticket>
2227 Remove the specified ticket from the current configuration.
2228 config [<node>]
2230 Show booth configuration from the specified node or from the
2231 current node if node not specified.
2232 create ip <address>
2234 Make the cluster run booth service on the specified ip address
2235 as a cluster resource. Typically this is used to run booth
2236 site.
2237 clean-enable-authfile
2239 Remove 'enable-authfile' option from booth configuration. This
2240 is useful when upgrading from booth that required the option to
2241 be present to a new version which doesn't tolerate the option.
2242 delete Remove booth resources created by the "pcs booth create" com‐
2244 mand.
2245 remove Remove booth resources created by the "pcs booth create" com‐
2247 mand.
2248 restart
2250 Restart booth resources created by the "pcs booth create" com‐
2251 mand.
2252 ticket grant <ticket> [<site address>]
2254 Grant the ticket to the site specified by the address, hence to
2255 the booth formation this site is a member of. When this specifi‐
2256 cation is omitted, site address that has been specified with
2257 'pcs booth create' command is used. Specifying site address is
2258 therefore mandatory when running this command at a host in an
2259 arbitrator role.
2260 Note that the ticket must not be already granted in given booth
2261 formation; for an ad-hoc (and, in the worst case, abrupt, for a
2262 lack of a direct atomicity) change of this preference baring di‐
2263 rect interventions at the sites, the ticket needs to be revoked
2264 first, only then it can be granted at another site again.
2265 ticket revoke <ticket> [<site address>]
2267 Revoke the ticket in the booth formation as identified with one
2268 of its member sites specified by the address. When this specifi‐
2269 cation is omitted, site address that has been specified with a
2270 prior 'pcs booth create' command is used. Specifying site ad‐
2271 dress is therefore mandatory when running this command at a host
2272 in an arbitrator role.
2273 status Print current status of booth on the local node.
2275 pull <node>
2277 Pull booth configuration from the specified node.
2278 sync [--skip-offline]
2280 Send booth configuration from the local node to all nodes in the
2281 cluster.
2282 enable Enable booth arbitrator service.
2284 disable
2286 Disable booth arbitrator service.
2287 start Start booth arbitrator service.
2289 stop Stop booth arbitrator service.
2291 status
2293 [status] [--full] [--hide-inactive]
2294 View all information about the cluster and resources (--full
2295 provides more details, --hide-inactive hides inactive re‐
2296 sources).
2297 resources [<resource id | tag id>] [node=<node>] [--hide-inactive]
2299 Show status of all currently configured resources. If --hide-in‐
2300 active is specified, only show active resources. If a resource
2301 or tag id is specified, only show status of the specified re‐
2302 source or resources in the specified tag. If node is specified,
2303 only show status of resources configured for the specified node.
2304 cluster
2306 View current cluster status.
2307 corosync
2309 View current membership information as seen by corosync.
2310 quorum View current quorum status.
2312 qdevice <device model> [--full] [<cluster name>]
2314 Show runtime status of specified model of quorum device
2315 provider. Using --full will give more detailed output. If
2316 <cluster name> is specified, only information about the speci‐
2317 fied cluster will be displayed.
2318 booth Print current status of booth on the local node.
2320 nodes [corosync | both | config]
2322 View current status of nodes from pacemaker. If 'corosync' is
2323 specified, view current status of nodes from corosync instead.
2324 If 'both' is specified, view current status of nodes from both
2325 corosync & pacemaker. If 'config' is specified, print nodes from
2326 corosync & pacemaker configuration.
2327 pcsd [<node>]...
2329 Show current status of pcsd on nodes specified, or on all nodes
2330 configured in the local cluster if no nodes are specified.
2331 xml View xml version of status (output from crm_mon -r -1 -X).
2333 config
2335 [show] View full cluster configuration.
2336 backup [filename]
2338 Creates the tarball containing the cluster configuration files.
2339 If filename is not specified the standard output will be used.
2340 restore [--local] [filename]
2342 Restores the cluster configuration files on all nodes from the
2343 backup. If filename is not specified the standard input will be
2344 used. If --local is specified only the files on the current
2345 node will be restored.
2346 checkpoint
2348 List all available configuration checkpoints.
2349 checkpoint view <checkpoint_number>
2351 Show specified configuration checkpoint.
2352 checkpoint diff <checkpoint_number> <checkpoint_number>
2354 Show differences between the two specified checkpoints. Use
2355 checkpoint number 'live' to compare a checkpoint to the current
2356 live configuration.
2357 checkpoint restore <checkpoint_number>
2359 Restore cluster configuration to specified checkpoint.
2360 pcsd
2362 certkey <certificate file> <key file>
2363 Load custom certificate and key files for use in pcsd.
2364 status [<node>]...
2366 Show current status of pcsd on nodes specified, or on all nodes
2367 configured in the local cluster if no nodes are specified.
2368 sync-certificates
2370 Sync pcsd certificates to all nodes in the local cluster.
2371 deauth [<token>]...
2373 Delete locally stored authentication tokens used by remote sys‐
2374 tems to connect to the local pcsd instance. If no tokens are
2375 specified all tokens will be deleted. After this command is run
2376 other nodes will need to re-authenticate against this node to be
2377 able to connect to it.
2378 host
2380 auth (<host name> [addr=<address>[:<port>]])... [-u <username>] [-p
2381 <password>]
2382 Authenticate local pcs/pcsd against pcsd on specified hosts. It
2383 is possible to specify an address and a port via which pcs/pcsd
2384 will communicate with each host. If an address is not specified
2385 a host name will be used. If a port is not specified 2224 will
2386 be used.
2387 deauth [<host name>]...
2389 Delete authentication tokens which allow pcs/pcsd on the current
2390 system to connect to remote pcsd instances on specified host
2391 names. If the current system is a member of a cluster, the to‐
2392 kens will be deleted from all nodes in the cluster. If no host
2393 names are specified all tokens will be deleted. After this com‐
2394 mand is run this node will need to re-authenticate against other
2395 nodes to be able to connect to them.
2396 node
2398 attribute [[<node>] [--name <name>] | <node> <name>=<value> ...]
2399 Manage node attributes. If no parameters are specified, show
2400 attributes of all nodes. If one parameter is specified, show
2401 attributes of specified node. If --name is specified, show
2402 specified attribute's value from all nodes. If more parameters
2403 are specified, set attributes of specified node. Attributes can
2404 be removed by setting an attribute without a value.
2405 maintenance [--all | <node>...] [--wait[=n]]
2407 Put specified node(s) into maintenance mode, if no nodes or op‐
2408 tions are specified the current node will be put into mainte‐
2409 nance mode, if --all is specified all nodes will be put into
2410 maintenance mode. If --wait is specified, pcs will wait up to
2411 'n' seconds for the node(s) to be put into maintenance mode and
2412 then return 0 on success or 1 if the operation not succeeded
2413 yet. If 'n' is not specified it defaults to 60 minutes.
2414 unmaintenance [--all | <node>...] [--wait[=n]]
2416 Remove node(s) from maintenance mode, if no nodes or options are
2417 specified the current node will be removed from maintenance
2418 mode, if --all is specified all nodes will be removed from main‐
2419 tenance mode. If --wait is specified, pcs will wait up to 'n'
2420 seconds for the node(s) to be removed from maintenance mode and
2421 then return 0 on success or 1 if the operation not succeeded
2422 yet. If 'n' is not specified it defaults to 60 minutes.
2423 standby [--all | <node>...] [--wait[=n]]
2425 Put specified node(s) into standby mode (the node specified will
2426 no longer be able to host resources), if no nodes or options are
2427 specified the current node will be put into standby mode, if
2428 --all is specified all nodes will be put into standby mode. If
2429 --wait is specified, pcs will wait up to 'n' seconds for the
2430 node(s) to be put into standby mode and then return 0 on success
2431 or 1 if the operation not succeeded yet. If 'n' is not specified
2432 it defaults to 60 minutes.
2433 unstandby [--all | <node>...] [--wait[=n]]
2435 Remove node(s) from standby mode (the node specified will now be
2436 able to host resources), if no nodes or options are specified
2437 the current node will be removed from standby mode, if --all is
2438 specified all nodes will be removed from standby mode. If --wait
2439 is specified, pcs will wait up to 'n' seconds for the node(s) to
2440 be removed from standby mode and then return 0 on success or 1
2441 if the operation not succeeded yet. If 'n' is not specified it
2442 defaults to 60 minutes.
2443 utilization [[<node>] [--name <name>] | <node> <name>=<value> ...]
2445 Add specified utilization options to specified node. If node is
2446 not specified, shows utilization of all nodes. If --name is
2447 specified, shows specified utilization value from all nodes. If
2448 utilization options are not specified, shows utilization of
2449 specified node. Utilization option should be in format
2450 name=value, value has to be integer. Options may be removed by
2451 setting an option without a value. Example: pcs node utiliza‐
2452 tion node1 cpu=4 ram= For the utilization configuration to be
2453 in effect, cluster property 'placement-strategy' must be config‐
2454 ured accordingly.
2455 alert
2457 [config]
2458 Show all configured alerts.
2459 create path=<path> [id=<alert-id>] [description=<description>] [options
2461 [<option>=<value>]...] [meta [<meta-option>=<value>]...]
2462 Define an alert handler with specified path. Id will be automat‐
2463 ically generated if it is not specified.
2464 update <alert-id> [path=<path>] [description=<description>] [options
2466 [<option>=<value>]...] [meta [<meta-option>=<value>]...]
2467 Update an existing alert handler with specified id. Unspecified
2468 options will be kept unchanged. If you wish to remove an option,
2469 set it to empty value, i.e. 'option_name='.
2470 delete <alert-id> ...
2472 Remove alert handlers with specified ids.
2473 remove <alert-id> ...
2475 Remove alert handlers with specified ids.
2476 recipient add <alert-id> value=<recipient-value> [id=<recipient-id>]
2478 [description=<description>] [options [<option>=<value>]...] [meta
2479 [<meta-option>=<value>]...]
2480 Add new recipient to specified alert handler.
2481 recipient update <recipient-id> [value=<recipient-value>] [descrip‐
2483 tion=<description>] [options [<option>=<value>]...] [meta [<meta-op‐
2484 tion>=<value>]...]
2485 Update an existing recipient identified by its id. Unspecified
2486 options will be kept unchanged. If you wish to remove an option,
2487 set it to empty value, i.e. 'option_name='.
2488 recipient delete <recipient-id> ...
2490 Remove specified recipients.
2491 recipient remove <recipient-id> ...
2493 Remove specified recipients.
2494 client
2496 local-auth [<pcsd-port>] [-u <username>] [-p <password>]
2497 Authenticate current user to local pcsd. This is required to run
2498 some pcs commands which may require permissions of root user
2499 such as 'pcs cluster start'.
2500 dr
2502 config Display disaster-recovery configuration from the local node.
2503 status [--full] [--hide-inactive]
2505 Display status of the local and the remote site cluster (--full
2506 provides more details, --hide-inactive hides inactive re‐
2507 sources).
2508 set-recovery-site <recovery site node>
2510 Set up disaster-recovery with the local cluster being the pri‐
2511 mary site. The recovery site is defined by a name of one of its
2512 nodes.
2513 destroy
2515 Permanently destroy disaster-recovery configuration on all
2516 sites.
2517 tag
2519 [config|list [<tag id>...]]
2520 Display configured tags.
2521 create <tag id> <id> [<id>]...
2523 Create a tag containing the specified ids.
2524 delete <tag id>...
2526 Delete specified tags.
2527 remove <tag id>...
2529 Delete specified tags.
2530 update <tag id> [add <id> [<id>]... [--before <id> | --after <id>]]
2532 [remove <id> [<id>]...]
2533 Update a tag using the specified ids. Ids can be added, removed
2534 or moved in a tag. You can use --before or --after to specify
2535 the position of the added ids relatively to some id already ex‐
2536 isting in the tag. By adding ids to a tag they are already in
2537 and specifying --after or --before you can move the ids in the
2538 tag.
2539
2541 Show all resources
2542 # pcs resource config
2543 Show options specific to the 'VirtualIP' resource
2545 # pcs resource config VirtualIP
2546 Create a new resource called 'VirtualIP' with options
2548 # pcs resource create VirtualIP ocf:heartbeat:IPaddr2
2549 ip=192.168.0.99 cidr_netmask=32 nic=eth2 op monitor interval=30s
2550 Create a new resource called 'VirtualIP' with options
2552 # pcs resource create VirtualIP IPaddr2 ip=192.168.0.99
2553 cidr_netmask=32 nic=eth2 op monitor interval=30s
2554 Change the ip address of VirtualIP and remove the nic option
2556 # pcs resource update VirtualIP ip=192.168.0.98 nic=
2557 Delete the VirtualIP resource
2559 # pcs resource delete VirtualIP
2560 Create the MyStonith stonith fence_virt device which can fence host
2562 'f1'
2563 # pcs stonith create MyStonith fence_virt pcmk_host_list=f1
2564 Set the stonith-enabled property to false on the cluster (which dis‐
2566 ables stonith)
2567 # pcs property set stonith-enabled=false
2568
2570 Various pcs commands accept the --force option. Its purpose is to over‐
2571 ride some of checks that pcs is doing or some of errors that may occur
2572 when a pcs command is run. When such error occurs, pcs will print the
2573 error with a note it may be overridden. The exact behavior of the op‐
2574 tion is different for each pcs command. Using the --force option can
2575 lead into situations that would normally be prevented by logic of pcs
2576 commands and therefore its use is strongly discouraged unless you know
2577 what you are doing.
2578
2580 EDITOR
2581 Path to a plain-text editor. This is used when pcs is requested
2582 to present a text for the user to edit.
2583 no_proxy, https_proxy, all_proxy, NO_PROXY, HTTPS_PROXY, ALL_PROXY
2585 These environment variables (listed according to their priori‐
2586 ties) control how pcs handles proxy servers when connecting to
2587 cluster nodes. See curl(1) man page for details.
2588
2590 This section summarizes the most important changes in commands done in
2591 pcs-0.11.x compared to pcs-0.10.x. For detailed description of current
2592 commands see above.
2593 Legacy role names
2595 Roles 'Master' and 'Slave' are deprecated and should not be used any‐
2596 more. Instead use 'Promoted' and 'Unpromoted' respectively. Similarly,
2597 --master has been deprecated and replaced with --promoted.
2598 cluster
2600 uidgid rm
2601 This command has been replaced with 'pcs cluster uidgid delete'
2602 and 'pcs cluster uidgid remove'.
2603 resource
2605 move The 'pcs resource move' now automatically removes location con‐
2606 straint used for moving a resource. It is equivalent of 'pcs re‐
2607 source move --autodelete' from pcs-0.10.9. Legacy functionality
2608 of the 'resource move' command is still available as 'resource
2609 move-with-constraint <resource id>'.
2610 show --full
2612 This command has been replaced with 'pcs resource config'.
2613 show --groups
2615 This command has been replaced with 'pcs resource group list'.
2616 show This command has been replaced with 'pcs resource status'.
2618 stonith
2620 show --full
2621 This command has been replaced with 'pcs stonith config'.
2622 show This command has been replaced with 'pcs stonith status'.
2624
2626 This section summarizes the most important changes in commands done in
2627 pcs-0.10.x compared to pcs-0.9.x. For detailed description of current
2628 commands see above.
2629 acl
2631 show The 'pcs acl show' command has been deprecated and will be re‐
2632 moved. Please use 'pcs acl config' instead. Applicable in
2633 pcs-0.10.9 and newer.
2634 alert
2636 show The 'pcs alert show' command has been deprecated and will be re‐
2637 moved. Please use 'pcs alert config' instead. Applicable in
2638 pcs-0.10.9 and newer.
2639 cluster
2641 auth The 'pcs cluster auth' command only authenticates nodes in a lo‐
2642 cal cluster and does not accept a node list. The new command for
2643 authentication is 'pcs host auth'. It allows one to specify host
2644 names, addresses and pcsd ports.
2645 node add
2647 Custom node names and Corosync 3.x with knet are fully supported
2648 now, therefore the syntax has been completely changed.
2649 The --device and --watchdog options have been replaced with 'de‐
2650 vice' and 'watchdog' options, respectively.
2651 pcsd-status
2653 The 'pcs cluster pcsd-status' command has been deprecated and
2654 will be removed. Please use 'pcs pcsd status' or 'pcs status
2655 pcsd' instead. Applicable in pcs-0.10.9 and newer.
2656 quorum This command has been replaced with 'pcs quorum'.
2658 remote-node add
2660 This command has been replaced with 'pcs cluster node
2661 add-guest'.
2662 remote-node remove
2664 This command has been replaced with 'pcs cluster node
2665 delete-guest' and its alias 'pcs cluster node remove-guest'.
2666 setup Custom node names and Corosync 3.x with knet are fully supported
2668 now, therefore the syntax has been completely changed.
2669 The --name option has been removed. The first parameter of the
2670 command is the cluster name now.
2671 The --local option has been replaced with --corosync_conf
2672 <path>.
2673 standby
2675 This command has been replaced with 'pcs node standby'.
2676 uidgid rm
2678 This command has been deprecated, use 'pcs cluster uidgid
2679 delete' or 'pcs cluster uidgid remove' instead.
2680 unstandby
2682 This command has been replaced with 'pcs node unstandby'.
2683 verify The -V option has been replaced with --full.
2685 To specify a filename, use the -f option.
2686 constraint
2688 list The 'pcs constraint list' command, as well as its variants 'pcs
2689 constraint [location | colocation | order | ticket] list', has
2690 been deprecated and will be removed. Please use 'pcs constraint
2691 [location | colocation | order | ticket] config' instead. Appli‐
2692 cable in pcs-0.10.9 and newer.
2693 show The 'pcs constraint show' command, as well as its variants 'pcs
2695 constraint [location | colocation | order | ticket] show', has
2696 been deprecated and will be removed. Please use 'pcs constraint
2697 [location | colocation | order | ticket] config' instead. Appli‐
2698 cable in pcs-0.10.9 and newer.
2699 pcsd
2701 clear-auth
2702 This command has been replaced with 'pcs host deauth' and 'pcs
2703 pcsd deauth'.
2704 property
2706 list The 'pcs property list' command has been deprecated and will be
2707 removed. Please use 'pcs property config' instead. Applicable in
2708 pcs-0.10.9 and newer.
2709 set The --node option is no longer supported. Use the 'pcs node at‐
2711 tribute' command to set node attributes.
2712 show The --node option is no longer supported. Use the 'pcs node at‐
2714 tribute' command to view node attributes.
2715 The 'pcs property show' command has been deprecated and will be
2716 removed. Please use 'pcs property config' instead. Applicable in
2717 pcs-0.10.9 and newer.
2718 unset The --node option is no longer supported. Use the 'pcs node at‐
2720 tribute' command to unset node attributes.
2721 resource
2723 create The 'master' keyword has been changed to 'promotable'.
2724 failcount reset
2726 The command has been removed as 'pcs resource cleanup' is doing
2727 exactly the same job.
2728 master This command has been replaced with 'pcs resource promotable'.
2730 show Previously, this command displayed either status or configura‐
2732 tion of resources depending on the parameters specified. This
2733 was confusing, therefore the command was replaced by several new
2734 commands. To display resources status, run 'pcs resource' or
2735 'pcs resource status'. To display resources configuration, run
2736 'pcs resource config' or 'pcs resource config <resource name>'.
2737 To display configured resource groups, run 'pcs resource group
2738 list'.
2739 status
2741 groups This command has been replaced with 'pcs resource group list'.
2742 stonith
2744 level add | clear | delete | remove
2745 Delimiting stonith devices with a comma is deprecated, use a
2746 space instead. Applicable in pcs-0.10.9 and newer.
2747 level clear
2749 Syntax of the command has been fixed so that it is not ambiguous
2750 any more. New syntax is 'pcs stonith level clear [target <tar‐
2751 get> | stonith <stonith id>...]'. Old syntax 'pcs stonith level
2752 clear [<target> | <stonith ids>]' is deprecated but still func‐
2753 tional in pcs-0.10.x. Applicable in pcs-0.10.9 and newer.
2754 level delete | remove
2756 Syntax of the command has been fixed so that it is not ambiguous
2757 any more. New syntax is 'pcs stonith level delete | remove [tar‐
2758 get <target>] [stonith <stonith id>]...'. Old syntax 'pcs
2759 stonith level delete | remove [<target>] [<stonith id>]...' is
2760 deprecated but still functional in pcs-0.10.x. Applicable in
2761 pcs-0.10.9 and newer.
2762 sbd device setup
2764 The --device option has been replaced with the 'device' option.
2765 sbd enable
2767 The --device and --watchdog options have been replaced with 'de‐
2768 vice' and 'watchdog' options, respectively.
2769 show Previously, this command displayed either status or configura‐
2771 tion of stonith resources depending on the parameters specified.
2772 This was confusing, therefore the command was replaced by sev‐
2773 eral new commands. To display stonith resources status, run 'pcs
2774 stonith' or 'pcs stonith status'. To display stonith resources
2775 configuration, run 'pcs stonith config' or 'pcs stonith config
2776 <stonith name>'.
2777 tag
2779 list The 'pcs tag list' command has been deprecated and will be re‐
2780 moved. Please use 'pcs tag config' instead. Applicable in
2781 pcs-0.10.9 and newer.
2782
2784 http://clusterlabs.org/doc/
2785 pcsd(8), pcs_snmp_agent(8)
2787 corosync_overview(8), votequorum(5), corosync.conf(5), corosync-qde‐
2789 vice(8), corosync-qdevice-tool(8), corosync-qnetd(8),
2790 corosync-qnetd-tool(8)
2791 pacemaker-controld(7), pacemaker-fenced(7), pacemaker-schedulerd(7),
2793 crm_mon(8), crm_report(8), crm_simulate(8)
2794 boothd(8), sbd(8)
2796pcs 0.11.6 2023-06-20 PCS(8)